A home decoration design method based on artificial intelligence

Abstract

The present invention discloses a home decoration design method based on artificial intelligence, which is applied to the field of home decoration design. The present invention comprises: S1, obtaining a variety of home decoration style data, and preprocessing the various home decoration style data; S2, dividing the various home decoration style data preprocessed in S1 into a training set and a test set; S3, establishing a home design model based on an AI algorithm, inputting the training set and the test set in S2 into the home design model to train and test the home design model; S4, inputting the user's preferences, needs, and room-related images into the trained home design model, and the model generates a variety of design schemes for the user to choose from. The present invention makes the generation of design schemes faster and more efficient, and can quickly generate a variety of design schemes for the user to choose from.

Description

Technical Field

[0001] The present invention relates to the technical field of home decoration design, and more specifically to a home decoration design method based on artificial intelligence. Background Art

[0002] With the development of the real estate industry, the demand for interior home design is also increasing. Due to the scarcity of excellent designer resources, the role of home design software with a high degree of intelligence in improving traffic transaction rate and consumer satisfaction is becoming increasingly prominent. The application of AI technology in the field of home design is in response to the needs of the "whole home customization" era. As the main application of the future intelligent interconnection system in the home field, the smart home system is closely related to the development of modern artificial intelligence technology.

[0003] Early designers mainly relied on traditional design tools and methods, such as hand-drawn sketches and CAD software, which have limitations in efficiency and accuracy. The level of intelligence in traditional home design is low, and many design and decision-making processes require human participation and judgment, which is time-consuming and labor-intensive. Personalized design is often limited by the designer's personal experience and creativity, and the design creativity is limited, making it difficult to meet the personalized needs of all users. Early technical limitations led to limited applications of AI in interior design and relatively single functions.

[0004] Therefore, it is an urgent problem for technicians in this field to propose a home decoration design method based on artificial intelligence to solve the difficulties existing in the prior art. Summary of the invention

[0005] In view of this, the present invention provides a home decoration design method based on artificial intelligence, which makes the generation of design schemes more rapid and efficient, and can quickly generate a variety of design schemes for users to choose from.

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

[0007] A home decoration design method based on artificial intelligence, comprising the following steps:

[0008] S1. Acquire a variety of home decoration style data, and pre-process the multiple home decoration style data;

[0009] S2, dividing the various home decoration style data preprocessed in S1 into a training set and a test set;

[0010] S3, establish a home design model based on AI algorithm, input the training set and test set in S2 into the home design model to train and test the home design model;

[0011] S4. Input the user's preferences, needs, and room-related images into the trained home design model, and the model generates multiple design options for the user to choose from.

[0012] Optionally, the various home decoration style data in S1 are image data, and preprocessing of the image data includes image cropping, scaling, denoising, and normalization.

[0013] Optionally, the home design model in S3 uses a generative adversarial network algorithm to process image data of various home decoration styles.

[0014] Optionally, the training of the home design model in S3 includes:

[0015] A generative adversarial network is used to extract design style features and transfer these features to new designs. A discriminator that understands the design style and a generator that generates images that meet user needs are trained. The home design model is trained using preprocessed image data. During the training process, the generator and discriminator continue to compete until the generator generates realistic design images.

[0016] Optionally, S3 evaluates the performance of the model by inputting the test set into the home design model. The performance of the model includes the quality, diversity and matching degree of the generated images with the user needs; the model is adjusted and optimized according to the evaluation results.

[0017] Optionally, the room-related image input by the user in S4 is a three-dimensional view containing measurement information generated by AR technology of a preset AI system.

[0018] It can be seen from the above technical solutions that, compared with the prior art, the present invention discloses a home decoration design method based on artificial intelligence, which has the following beneficial effects:

[0019] The present invention uses AI technology to make the generation of design solutions faster and more efficient, and can quickly generate a variety of design solutions for users to choose from; uses a generative adversarial network machine learning algorithm to build a home design model, and generates new design solutions with higher creativity to meet the different personalized needs of users; uses computer vision recognition technology to identify the space, and after identifying the space, uses 3D modeling and rendering technology to create a virtual model or animation, providing users with a more intuitive home design experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0021] Figure 1 A flow chart of a home decoration design method based on artificial intelligence provided by the present invention. DETAILED DESCRIPTION

[0022] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0023] Reference Figure 1 As shown, the present invention discloses a home decoration design method based on artificial intelligence, comprising the following steps:

[0024] S1. Acquire a variety of home decoration style data, and pre-process the multiple home decoration style data;

[0025] S2, dividing the various home decoration style data preprocessed in S1 into a training set and a test set;

[0026] S3, establish a home design model based on AI algorithm, input the training set and test set in S2 into the home design model to train and test the home design model;

[0027] S4. Input the user's preferences, needs, and room-related images into the trained home design model, and the model generates multiple design options for the user to choose from.

[0028] Furthermore, the various home decoration style data in S1 are image data, and the preprocessing of the image data includes image cropping, scaling, denoising, and normalization.

[0029] Furthermore, the home design model in S3 uses a generative adversarial network algorithm to process image data of various home decoration styles.

[0030] Furthermore, the training of the home design model in S3 includes:

[0031] Generative adversarial networks are used to extract design style features and transfer these features to new designs. A discriminator that understands design style and a generator that generates images that meet user needs are trained. The home design model is trained using preprocessed image data. During the training process, the generator and discriminator continue to compete until the generator can generate realistic design images.

[0032] Furthermore, S3 evaluates the performance of the model by inputting the test set into the home design model. The performance of the model includes the quality, diversity and matching degree of the generated images with user needs; the model is adjusted and optimized according to the evaluation results.

[0033] Furthermore, the room-related image input by the user in S4 is a three-dimensional view containing measurement information generated by the AR technology of the preset AI system.

[0034] Specifically, users need to provide room dimensions, shape, and any existing decorative or architectural features. The AI ​​system can use AR (augmented reality) technology to help users quickly measure and obtain accurate spatial data.

[0035] 1. Demand analysis: clarify the user's needs, including the type of space to be measured, size range, accuracy requirements, etc.

[0036] 2. Data acquisition: Use the camera of an AI system (such as a smartphone or tablet) to capture images or videos of the real-world environment. Move the device around in space to capture different perspectives and information.

[0037] 3. Feature extraction: Use computer vision algorithms to extract key features from the collected data, such as edges, corners, planes, etc. These features are helpful for subsequent spatial modeling and measurement.

[0038] 4. Spatial modeling: Based on the extracted features, a 3D model of the space is constructed using 3D reconstruction technology.

[0039] 5. Dimension measurement: Perform dimensional measurements in 3D models, such as calculating distances, angles, areas, etc. This can be done using AI algorithms to automatically identify and measure, or by users manually specifying measurement points and lines.

[0040] 6. Augmented reality presentation: The measurement results are superimposed on the real-time image in the form of virtual objects, providing users with an intuitive augmented reality experience.

[0041] 7. Result output: The measurement results are presented to the user in digital form, generating an augmented reality view containing the measurement information.

[0042] Specifically, the steps for generating multiple design solutions by the home design model in S4 are as follows:

[0043] 1. Content generation: The model predicts the probability distribution of subsequent design elements based on the user’s input preferences, requirements, and the 3D view of the room.

[0044] 2. Feature extraction and style transfer: Use generative adversarial networks to extract design style features and transfer these features to new designs. This may involve training a discriminator network that can understand the design style and a generator network that can generate images that meet user needs.

[0045] 3. Scheme generation: Using the trained generative design model, a variety of possible design schemes are generated based on the design parameters input by the user. These schemes vary in style, layout, color, etc. to provide diverse choices.

[0046] 4. Scheme evaluation and selection: Evaluate the generated design schemes and select the schemes that meet the user's preferences and needs. The evaluation can be based on preset design principles, user feedback or other heuristic methods; each scheme may include wall color, floor material, furniture layout, lighting design and decorations, etc.

[0047] After the user selects a satisfactory preliminary design plan, the AI ​​system further refines the design details, such as material selection, texture, decorations, etc.; the AI ​​system uses optimization algorithms and genetic algorithms to find the optimal furniture placement plan based on the dimensional measurement results of the 3D model and the actual size of the room, the size and shape of the furniture, and other factors; the AI ​​system uses 3D modeling software to create a virtual model of the design plan so that the user can see the final effect more intuitively.

[0048] 3D simulation: Use 3D modeling and rendering technology to create virtual models or animations, allowing users to experience the design effects more intuitively.

[0049] a) Based on the layout optimization of the preliminary design plan, use 3D modeling software to build a three-dimensional model of the space and furniture. Describe the size, shape and material properties of each object in detail to ensure the accuracy and realism of the model.

[0050] b) Set materials and textures: Give the model appropriate materials and textures, such as wood grain on the floor and paint texture on the wall, to enhance the realism of the visual effect. Adjust the material's reflectivity, transparency, roughness and other properties to simulate the appearance under different lighting conditions.

[0051] c) Arrange light sources and cameras: Add light sources such as natural light, point light sources or spotlights to the scene to simulate real-life lighting conditions. Set the position and angle of the camera to define the user's viewing angle.

[0052] d) Rendering the scene: Use a rendering engine to render the scene, calculate how light interacts with objects and casts shadows, and generate high-quality static images or dynamic videos. The rendering process may require adjusting rendering parameters such as resolution, sampling rate, and rendering channels to achieve the desired visual effect and rendering speed.

[0053] e) Post-processing: Post-process the rendered images or videos, such as color correction, brightness contrast adjustment, special effects addition, etc., to enhance the visual presentation. Virtual characters or dynamic elements can be added to make the scene more vivid and convincing.

[0054] f) User interactive experience: Display rendered images or animations to users through appropriate platforms, such as computers, tablets, or VR devices. For virtual reality (VR) or augmented reality (AR) experience.

[0055] g) Feedback and Adjustment: Collect user feedback on the 3D model, adjust the design and 3D model based on user feedback, re-render and provide an updated experience.

[0056] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0057] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A home decoration design method based on artificial intelligence, characterized in that: The following steps are involved: S1. Acquire a variety of home decoration style data, and pre-process the multiple home decoration style data; S2, dividing the various home decoration style data preprocessed in S1 into a training set and a test set; S3, establish a home design model based on AI algorithm, input the training set and test set in S2 into the home design model to train and test the home design model; S4, inputting the user's preferences, needs, and room-related images into the trained home design model, and the model generates multiple design options for the user to choose from; The room-related images input by the user in S4 are three-dimensional views containing measurement information generated by the preset AI system combined with AR technology; Specifically, based on the room size and shape provided by the user, as well as any existing decorative or architectural features, the AI ​​system uses AR technology to measure and obtain spatial data. The specific steps are as follows: Demand analysis: clarify the user's needs, including the type of space to be measured, size range, and accuracy requirements; Data acquisition: Use the AI ​​system’s camera to capture images or videos of the real world, moving the device around in space to capture different perspectives and information. Feature extraction: Use computer vision algorithms to extract key features from collected data, including edges, corners, and planes; Spatial modeling: Based on the extracted features, a 3D model of the space is constructed using 3D reconstruction technology; Dimension measurement: Dimension measurement in 3D models, using AI algorithms to automatically identify and measure distances, angles, and areas, or by manually specifying measurement points and lines by the user; Augmented reality presentation: Overlay the measurement results into the real-time image in the form of virtual objects, providing users with an intuitive augmented reality experience; Result output: The measurement results are presented to the user in digital form, generating an augmented reality view containing the measurement information; The specific steps for the model in S4 to generate multiple design solutions are: Content generation: The model predicts the probability distribution of subsequent design elements based on the user's input preferences, requirements, and the 3D view of the room; Feature extraction and style transfer: Use generative adversarial networks to extract design style features and transfer the design style features to new designs; Plan generation: Using the trained home design model, a variety of design plans with different styles, layouts, and colors are generated based on the design parameters input by the user; Scheme evaluation and selection: Evaluate the generated multiple design schemes based on preset design principles, user feedback or heuristic methods, and select preliminary design schemes that meet user preferences and needs; After the user selects a satisfactory preliminary design plan, the AI ​​system refines the design details including material selection, texture, and decorations. It uses optimization algorithms and genetic algorithms to find the optimal furniture placement plan based on the dimensional measurement results of the 3D model and the actual size of the room, as well as the size and shape of the furniture. It uses 3D modeling software to create a virtual model of the design plan so that the final effect can be seen intuitively.

2. The artificial intelligence-based home decoration design method according to claim 1, characterized in that: The various home decoration style data in S1 are image data, and the preprocessing of the image data includes image cropping, scaling, denoising, and normalization.

3. The home decoration design method based on artificial intelligence according to claim 1, characterized in that: The home design model in S3 uses a generative adversarial network algorithm to process image data of various home decoration styles.

4. The method for home decoration design based on artificial intelligence according to claim 3, characterized in that: The training of the home design model in S3 includes: A generative adversarial network is used to extract design style features and transfer these features to new designs. A discriminator that understands the design style and a generator that generates images that meet user needs are trained. The home design model is trained using preprocessed image data. During the training process, the generator and discriminator continue to compete until the generator generates realistic design images.

5. The artificial intelligence-based home decoration design method according to claim 1, characterized in that: In S3, the performance of the model is evaluated by inputting the test set into the home design model. The performance of the model includes the quality, diversity and matching degree of the generated images with user needs; the model is adjusted and optimized according to the evaluation results.

Images