WO2023124160A1

WO2023124160A1 - Method, system and apparatus for automatically generating three-dimensional house layout, and medium

Info

Publication number: WO2023124160A1
Application number: PCT/CN2022/115289
Authority: WO
Inventors: 谭明奎; 周佳楸; 吴祥淼; 陈奇; 吴琦; 李远清
Original assignee: 华南理工大学
Priority date: 2021-12-27
Filing date: 2022-08-26
Publication date: 2023-07-06
Also published as: CN114373054A

Abstract

Disclosed in the present invention are a method, system and apparatus for automatically generating a three-dimensional house layout, and a medium. The method comprises: acquiring a building structurization requirement of a user, wherein the building structurization requirement comprises a layout requirement and a contour requirement; parsing the layout requirement and the contour requirement on the basis of a building data set, and by means of searching, obtaining a building bubble diagram which meets the layout requirement and the contour requirement; acquiring a two-dimensional building layout according to the obtained building bubble diagram and the contour requirement; and performing three-dimensional rendering according to the two-dimensional building layout, so as to generate a three-dimensional building layout as an output three-dimensional building model. The present invention converts a user requirement into a simple structurization requirement, thereby reducing the redundancy of the user requirement, and reducing the burden for the user to provide requirements. The present invention can be widely applied to the technical field of intelligent identification.

Description

A method, system, device and medium for automatically generating a three-dimensional house layout

technical field

The invention relates to the technical field of intelligent identification, in particular to a method, system, device and medium for automatically generating a three-dimensional house layout.

Background technique

At present, in the field of architectural design, only professional architectural designers can design building layouts. For professional architects, layout design is a very time-consuming task, and non-professional users can , and cannot carry out architectural house design. For this reason, more and more automated building layout design methods are widely used. However, since the user is not a professional architectural designer, the user will always provide some redundant and complex requirements, some emotional requirements such as I hope to have a very warm house, which makes it difficult for the architect to implement direct design. Furthermore, the user's building layout generally provides a fixed layout outline, that is, a building site, so that the layout outline needs to be considered during the design process of the room layout. Generally speaking, room layouts cannot exceed the size of the layout outline. However, the irregular geometric shape of the layout outline itself makes it difficult to extract the feature information of the layout outline, and the problem of how to effectively use the architectural layout outline needs to be solved urgently.

To sum up, how to simplify user requirements and automatically generate house layouts in combination with irregular layout outlines is an urgent problem to be solved by those skilled in the art.

Contents of the invention

In order to solve one of the technical problems in the prior art at least to a certain extent, the object of the present invention is to provide a method, system, device and medium for automatically generating a three-dimensional house layout.

The technical scheme adopted in the present invention is:

A method for automatically generating a three-dimensional house layout, comprising the following steps:

Obtaining the user's architectural structural requirements, wherein the architectural structural requirements include layout requirements and outline requirements;

Analyzing the layout requirements and the outline requirements based on a building dataset, and searching for building bubble diagrams that meet the layout requirements and the outline requirements;

Obtaining a two-dimensional building layout according to the obtained building bubble diagram and the outline requirement;

Perform three-dimensional rendering according to the two-dimensional building layout to generate a three-dimensional building layout as an output three-dimensional building model.

Further, the architectural structural requirements also include style requirements;

The three-dimensional rendering according to the two-dimensional building layout is performed to generate a three-dimensional building layout, and the output three-dimensional building model includes:

Obtain style requirements, and convert the style requirements into decoration texture pictures for the floors and walls of each room;

Rendering the two-dimensional building layout by using the decoration texture image, and generating a three-dimensional model according to a preset height;

Retrieving a three-dimensional furniture model from a preset home database according to the style requirements;

The three-dimensional furniture model is mapped to the three-dimensional model to obtain a three-dimensional architectural layout as an output three-dimensional architectural model.

Further, the mapping the 3D furniture model to the 3D model includes:

mapping the 3D furniture model to the 3D model according to a 3D transformation matrix;

Among them, the expression of mapping is as follows:

In the formula,

Represents the 3D furniture model, which is composed of 3D coordinates; q _i represents the position of the 3D furniture model in the 3D model; T represents the 3D transformation matrix.

Further, the layout requirements include room types and room numbers for the room layout, and the outline requirements include the geometric outline of a given site;

The analyzing the layout requirements and the outline requirements based on the building data set, and searching for building bubble diagrams that meet the layout requirements and the outline requirements include:

Search in the building data set according to the layout requirements, and obtain the first building bubble map and outline pair data that meet the user's room type and room number;

performing outline sorting on the first building bubble map and the outline pair data according to the outline requirement, to obtain building bubble map data matching the outline requirement;

The building bubble map data is further adjusted according to the outline requirement to obtain a building bubble map meeting the layout requirement and the outline requirement.

Further, the obtaining a two-dimensional building layout according to the obtained building bubble diagram and the outline requirement includes:

Using a pre-trained two-dimensional layout feature generation model to perform feature extraction on the building bubble map and the outline requirements to obtain the bounding box of the room layout;

The bounding box of the room layout is fused according to the outline requirement to obtain a two-dimensional building layout.

Further, the two-dimensional layout feature generation model is obtained in the following manner:

Use the first linear layer and the first pooling layer to construct a graph neural network;

Construct a recurrent neural network using the second linear layer, linear rectification layer, second pooling layer, and batch normalization layer;

Constructing a two-dimensional layout feature generation model according to the graph neural network and the cyclic neural network, inserting a bounding box regression loss function into the two-dimensional layout feature generation model as a target loss function;

Combining the target loss function with a training set to train the two-dimensional layout feature generation model to obtain the trained two-dimensional layout feature generation model;

Among them, the training set includes building bubble map and contour data.

Further, the feature extraction of the building bubble map and the outline requirements by using the pre-trained two-dimensional layout feature generation model includes:

Converting the user's profile requirement into the form of a profile point set, extracting features of the profile through the constructed cyclic neural network;

Feature extraction is performed on the building bubble map through the graph neural network to obtain features of the building bubble map.

Another technical scheme adopted in the present invention is:

A system for automatically generating 3D house layouts, including:

A requirement acquisition module, configured to acquire the user's architectural structural requirements, wherein the architectural structural requirements include layout requirements and outline requirements;

A data search module, configured to analyze the layout requirements and the outline requirements based on a building dataset, and search for building bubble diagrams that meet the layout requirements and the outline requirements;

A two-dimensional layout module, configured to obtain a two-dimensional building layout according to the obtained building bubble diagram and the outline requirements;

The three-dimensional layout module is used to perform three-dimensional rendering according to the two-dimensional building layout, and generate a three-dimensional building layout as an output three-dimensional building model.

Another technical scheme adopted in the present invention is:

A device for automatically generating a three-dimensional house layout, comprising:

at least one processor;

at least one memory for storing at least one program;

When the at least one program is executed by the at least one processor, the at least one processor implements the above method.

Another technical scheme adopted in the present invention is:

A computer-readable storage medium stores a processor-executable program therein, and the processor-executable program is used to perform the above method when executed by a processor.

The beneficial effects of the present invention are: the present invention transforms user requirements into simple structured requirements, reduces the redundancy of user requirements, and lightens the burden on users to provide requirements.

Description of drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following describes the accompanying drawings of the embodiments of the present invention or the related technical solutions in the prior art. It should be understood that the accompanying drawings in the following introduction are only In order to clearly describe some embodiments of the technical solutions of the present invention, those skilled in the art can also obtain other drawings based on these drawings without creative work.

Fig. 1 is the implementation flowchart of a method for automatically generating a three-dimensional house layout in an embodiment of the present invention;

Fig. 2 is an algorithm frame diagram of a method for automatically generating a three-dimensional house layout in an embodiment of the present invention;

Fig. 3 is a demand analysis schematic diagram of a method for automatically generating a three-dimensional house layout in an embodiment of the present invention;

Fig. 4 is a two-dimensional house layout visualization result of a method for automatically generating a three-dimensional house layout in an embodiment of the present invention;

Fig. 5 is a visualization result of a three-dimensional house layout of a method for automatically generating a three-dimensional house layout in an embodiment of the present invention;

Fig. 6 is a visualization result of diversified 3D houses of a method for automatically generating a 3D house layout in an embodiment of the present invention.

Detailed ways

Embodiments of the present invention are described in detail below, examples of which are shown in the drawings, wherein the same or similar reference numerals designate the same or similar elements or elements having the same or similar functions throughout. The embodiments described below by referring to the figures are exemplary only for explaining the present invention and should not be construed as limiting the present invention. For the step numbers in the following embodiments, it is only set for the convenience of illustration and description, and the order between the steps is not limited in any way. The execution order of each step in the embodiments can be adapted according to the understanding of those skilled in the art sexual adjustment.

In the description of the present invention, it should be understood that the orientation descriptions, such as up, down, front, back, left, right, etc. indicated orientations or positional relationships are based on the orientations or positional relationships shown in the drawings, and are only In order to facilitate the description of the present invention and simplify the description, it does not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as limiting the present invention.

In the description of the present invention, several means one or more, and multiple means more than two. Greater than, less than, exceeding, etc. are understood as not including the original number, and above, below, within, etc. are understood as including the original number. If the description of the first and second is only for the purpose of distinguishing the technical features, it cannot be understood as indicating or implying the relative importance or implicitly indicating the number of the indicated technical features or implicitly indicating the order of the indicated technical features relation.

In the description of the present invention, unless otherwise clearly defined, words such as setting, installation, and connection should be understood in a broad sense, and those skilled in the art can reasonably determine the specific meanings of the above words in the present invention in combination with the specific content of the technical solution.

As shown in Figure 1, this embodiment provides a method for automatically generating a three-dimensional house layout, which automatically generates a three-dimensional house layout based on structured requirements. It is required to automatically generate the algorithm frame diagram of the 3D house layout. The present embodiment method comprises the following steps:

S1. Obtain the user's architectural structural requirements, wherein the architectural structural requirements include layout requirements and outline requirements.

In the process of obtaining user requirements, the method of this embodiment defines the user requirements as architectural structural requirements of layout requirements and outline requirements. Among them, the layout requirements include the user's requirements for the room type and the number of rooms in the building layout, and the outline requirements include the planning requirements for the building land, that is, the geometric outline of the given land for the user's layout.

S2. Analyze the layout requirements and outline requirements based on the building data set, and search for building bubble diagrams that meet the layout requirements and outline requirements.

Wherein, step S2 includes steps S21-S23:

S21. Search in the building data set according to the layout requirements, and obtain the first building bubble map and outline pair data satisfying the user's room type and room number;

S22. Perform outline sorting on the first building bubble map and outline data according to outline requirements, and obtain building bubble map data matching outline requirements;

S23. Further adjust the architectural bubble map data according to the outline requirements, and obtain an architectural bubble map that meets the layout requirements and outline requirements.

Referring to Fig. 3, Fig. 3 shows a schematic diagram of parsing layout and outline requirements. Based on the existing building data set, firstly, through the analysis of the layout requirements, that is, the number requirements of the living room, bedroom, bathroom, etc., the building bubble map and contour pairs that meet the number requirements are searched in the data set. In order to determine whether the searched architectural bubble map meets the contour requirements, the method of this embodiment compares the contour requirements with the contours in the searched contour pairs to sort the degree to which the architectural bubble diagrams meet the contour requirements. After obtaining the first few architectural bubble diagrams that meet the contour requirements, adjust the architectural bubble diagrams to meet the contour requirements.

S3. Obtain a two-dimensional building layout according to the obtained building bubble map and outline requirements.

Using the obtained building bubble map and outline requirements, use the neural network for feature processing, and post-processing to obtain the two-dimensional building layout. Optionally, step S3 includes steps S31-S32:

S31. Using the pre-trained two-dimensional layout feature generation model to perform feature extraction on the building bubble map and outline requirements, and obtain the bounding box of the room layout;

S32. Fusion processing is performed on the bounding box of the room layout according to the outline requirement to obtain a two-dimensional building layout.

In some embodiments, the two-dimensional layout feature generation model is obtained by:

A1. Use the first linear layer and the first pooling layer to construct a graph neural network;

A2. Construct a recurrent neural network using the second linear layer, linear rectification layer, second pooling layer, and batch normalization layer;

A3. Construct a two-dimensional layout feature generation model based on graph neural network and recurrent neural network, and insert a bounding box regression loss function into the two-dimensional layout feature generation model as the target loss function;

A4. Combine the target loss function with the building bubble map and outline data in the training set to train the two-dimensional layout feature generation model, and obtain the trained two-dimensional layout feature generation model.

Based on the above-mentioned two-dimensional layout feature generation model, step S31 includes steps S311-S312:

S311. Convert the user's profile requirement into a form of profile point set, and extract features of the profile through the constructed cyclic neural network;

S312. Perform feature extraction on the building bubble map through the graph neural network to obtain features of the building bubble map.

Through the building bubble map and outline requirements obtained in step S2, the method of this embodiment utilizes the linear layer and the pooling layer to construct a graph neural network to extract the characteristics of the building bubble map; using the linear layer, linear rectification layer, pooling layer, batch regression One layer builds a recurrent neural network. For the contour, it is processed into the form of a point set, and the contour features are extracted through the constructed cyclic neural network considering the order of the contour point set. Combining the features obtained by the graph neural network and the recurrent neural network, the overall features for predicting the layout of two-dimensional buildings are obtained. The method of this embodiment inserts the bounding box regression loss function as the target loss function:

where o _i denote the ground-truth labels of the 2D bounding boxes of each room layout,

Denotes the predicted 2D bounding boxes obtained using the general features of 2D building layouts, and M denotes the number of rooms in each building layout.

Using the two-dimensional bounding box obtained in the above method, the bounding box is further aligned, and the bounding box is constrained by the outline requirement, so as to obtain the two-dimensional building layout diagram.

S4. Perform 3D rendering according to the 2D building layout to generate a 3D building layout as an output 3D building model.

In some embodiments, the output 3D building layout is directly used as the 3D building model finally output to the user, that is, the 3D building model does not include decorations.

In some embodiments, after the three-dimensional building layout is arranged by calling the three-dimensional home model, the three-dimensional building model is finally output to the user. The specific implementation is as follows:

The architectural structural requirements also include style requirements, that is, users also need to input style requirements, and the style requirements include requirements for architectural decoration styles, such as European style and Nordic style. Optionally, step S4 includes steps S41-S44:

S41. Obtain style requirements, and transform the style requirements into decoration texture pictures for floors and walls of each room;

S42. Use the decoration texture image to render the two-dimensional building layout, and generate a three-dimensional model according to the preset height;

S43. Retrieving a three-dimensional furniture model from a preset home database according to style requirements;

S44. Map the 3D furniture model to a 3D model to obtain a 3D building layout as an output 3D building model.

According to the user's style requirements, the texture scheme of the floor and wall of the two-dimensional building layout is obtained, and a certain height is set to generate a three-dimensional building model. Then, use the 3D rendering tool to render the texture scheme into the 3D architectural model to obtain the rendered 3D architectural model. Then, by setting the furniture layout rules, the 3D furniture model is mapped to the 3D architectural model to obtain a 3D architectural model that meets the user's requirements.

In some optional embodiments, the 3D home model is mapped to the 3D building model in the following manner:

According to the style requirements, the furniture is automatically arranged. The method of this embodiment maps the pre-designed 3D furniture model to the 3D building layout through the 3D transformation matrix:

in

Represents the 3D furniture model, which is composed of 3D coordinates, q _i represents the position of the 3D furniture model in the 3D architectural layout, T represents the 3D transformation matrix, through the bounding box of the 3D furniture model and the bounding box of the 3D furniture model position in the 3D architectural layout predict.

In some optional embodiments, in the three-dimensional building layout, the method of this embodiment arranges the furniture by setting two rules, the first is the global rule, the newly placed furniture not only cannot interfere with the placed furniture, And it cannot interfere with the activity space of doors and windows; the second is specific furniture rules. Different furniture has different furniture placement rules. For example, for a bed, it needs to be placed in the middle of the wall, while the bedside table needs to be placed in the middle of the wall. sides of the bed.

The above method will be explained in detail below in combination with experimental comparison and accompanying drawings.

1) Experimental numerical results

The above-mentioned method of automatically generating 3D house layout based on structured requirements was applied to the building datasets RPLAN and T3HM. Compared with other methods, the intersection of the predicted 2D layout bounding box and the real 2D layout bounding box obtained by the method in this embodiment The results of the union ratio and the intersection ratio with the contour are shown in Table 1 below:

Table 1: Results on T3HM and RPLAN

In order to further verify the effectiveness of the method, the method of this embodiment has also carried out human research and investigation. Compared with other methods, the method of this embodiment has reached the best performance at present, and the results are shown in Table 2 below:

Table 2: Results of different methods using RPLAN for human research experiments

2) Visualize the results

To further illustrate the effectiveness of this method, the visualization results are shown in the figure below.

In Figure 4, based on the same needs of users, using our method together with other methods to generate 2D house layouts, we provide the visualization results of 2D building layout diagrams on the RPLAN dataset. From the results, we can see that our method can provide a more regular room layout.

In order to verify the effect of this method in generating 3D house layouts, we use our method and other methods to generate 2D house layouts based on user needs, and use our 3D rendering method for rendering. Figure 5 shows the visualization results of this method on 3D house layout generation. In addition, in order to verify that our method can generate different 3D house layouts, in Fig. 6, we provide the visualization results of diverse 3D house layouts.

To sum up, the method of this embodiment obtains the user's architectural structural requirements for layout, outline and style, and then searches the layout and outline requirements in the building data set to obtain building bubble map data that meets the user's layout and outline requirements. The feature of architectural bubble map data is extracted by graph neural network, and the profile data feature is extracted by recurrent neural network. Then the features of the two are combined to generate the bounding box of the room layout, and the bounding box is post-processed to obtain the two-dimensional building layout. Then, combined with the user's stylized needs and the two-dimensional building layout, the three-dimensional model is generated by using a fixed height, and the three-dimensional rendering and the arrangement of the three-dimensional furniture model can be used to obtain the three-dimensional building model. Compared with the prior art, the specific beneficial effects are as follows:

(1) Since user requirements are transformed into simple structured requirements, the redundancy of user requirements is reduced, and the burden on users to provide requirements is reduced.

(2) In the processing of contour data, since the contour data is converted into a point set and feature extraction is performed using a cyclic neural network, the amount of network calculations for contour data processing can be greatly reduced and improved. Therefore, the method of automatically generating a three-dimensional house layout based on structural requirements provided by the embodiment of the present invention can be applied to the field of architectural design.

This embodiment also provides a system for automatically generating a three-dimensional house layout, including:

A system for automatically generating a three-dimensional house layout in this embodiment can execute a method for automatically generating a three-dimensional house layout provided by the method embodiment of the present invention, can execute any combination of implementation steps of the method embodiment, and has the corresponding Functions and beneficial effects.

This embodiment also provides a device for automatically generating a three-dimensional house layout, including:

at least one processor;

at least one memory for storing at least one program;

When the at least one program is executed by the at least one processor, the at least one processor implements the method shown in FIG. 1 .

A device for automatically generating a three-dimensional house layout in this embodiment can execute a method for automatically generating a three-dimensional house layout provided by the method embodiment of the present invention, can execute any combination of implementation steps in the method embodiment, and has the corresponding Functions and beneficial effects.

The embodiment of the present application also discloses a computer program product or computer program, where the computer program product or computer program includes computer instructions, and the computer instructions are stored in a computer-readable storage medium. The processor of the computer device can read the computer instruction from the computer-readable storage medium, and the processor executes the computer instruction, so that the computer device executes the method shown in FIG. 1 .

This embodiment also provides a storage medium, which stores an instruction or program that can execute a method for automatically generating a three-dimensional house layout provided by the method embodiment of the present invention. When the instruction or program is executed, the method embodiment can be executed Any combination of implementation steps has the corresponding functions and beneficial effects of the method.

In some alternative implementations, the functions/operations noted in the block diagrams may occur out of the order noted in the operational diagrams. For example, two blocks shown in succession may, in fact, be executed substantially concurrently or the blocks may sometimes be executed in the reverse order, depending upon the functionality/operations involved. Furthermore, the embodiments presented and described in the flowcharts of the present invention are provided by way of example in order to provide a more comprehensive understanding of the technology. The disclosed methods are not limited to the operations and logical flow presented herein. Alternative embodiments are contemplated in which the order of various operations is changed and in which sub-operations described as part of larger operations are performed independently.

Furthermore, although the invention has been described in the context of functional modules, it should be understood that one or more of the described functions and/or features may be integrated into a single physical device and/or unless stated to the contrary. or software modules, or one or more functions and/or features may be implemented in separate physical devices or software modules. It will also be appreciated that a detailed discussion of the actual implementation of each module is not necessary to understand the present invention. Rather, given the attributes, functions and internal relationships of the various functional blocks in the devices disclosed herein, the actual implementation of the blocks will be within the ordinary skill of the engineer. Accordingly, those skilled in the art can implement the present invention set forth in the claims without undue experimentation using ordinary techniques. It is also to be understood that the particular concepts disclosed are illustrative only and are not intended to limit the scope of the invention which is to be determined by the appended claims and their full scope of equivalents.

If the functions described above are realized in the form of software function units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the essence of the technical solution of the present invention or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including Several instructions are used to make a computer device (which may be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage media include: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disk or optical disc, etc., which can store program codes. .

The logic and/or steps represented in the flowcharts or otherwise described herein, for example, can be considered as a sequenced listing of executable instructions for implementing logical functions, can be embodied in any computer-readable medium, For use with instruction execution systems, devices, or devices (such as computer-based systems, systems including processors, or other systems that can fetch instructions from instruction execution systems, devices, or devices and execute instructions), or in conjunction with these instruction execution systems, devices or equipment for use. For the purposes of this specification, a "computer-readable medium" may be any device that can contain, store, communicate, propagate or transmit a program for use in or in conjunction with an instruction execution system, device or device.

More specific examples (non-exhaustive list) of computer-readable media include the following: electrical connection with one or more wires (electronic device), portable computer disk case (magnetic device), random access memory (RAM), Read Only Memory (ROM), Erasable and Editable Read Only Memory (EPROM or Flash Memory), Fiber Optic Devices, and Portable Compact Disc Read Only Memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program can be printed, as it may be possible, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or other suitable processing if necessary. The program is processed electronically and stored in computer memory.

It should be understood that various parts of the present invention can be realized by hardware, software, firmware or their combination. In the embodiments described above, various steps or methods may be implemented by 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 by any one or combination of the following techniques known in the art: Discrete logic circuits, ASICs with suitable combinational logic gates, programmable gate arrays (PGAs), field programmable gate arrays (FPGAs), etc.

In the above description of this specification, the description with reference to the terms "one embodiment/example", "another embodiment/example" or "some embodiments/example" means that the description is described in conjunction with the embodiment or example. A particular feature, structure, material, or characteristic is included in at least one embodiment or example of the invention. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the described specific features, structures, materials or characteristics may be combined in any suitable manner in any one or more embodiments or examples.

Although the embodiments of the present invention have been shown and described, those skilled in the art can understand that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principle and spirit of the present invention. The scope of the invention is defined by the claims and their equivalents.

The above is a specific description of the preferred implementation of the present invention, but the present invention is not limited to the above-mentioned embodiments, and those skilled in the art can also make various equivalent deformations or replacements without violating the spirit of the present invention. Equivalent modifications or replacements are all within the scope defined by the claims of the present application.

Claims

A method for automatically generating a three-dimensional house layout, comprising the following steps:

Obtaining the user's architectural structural requirements, wherein the architectural structural requirements include layout requirements and outline requirements;

Analyzing the layout requirements and the outline requirements based on a building dataset, and searching for building bubble diagrams that meet the layout requirements and the outline requirements;

Obtaining a two-dimensional building layout according to the obtained building bubble diagram and the outline requirement;

Perform three-dimensional rendering according to the two-dimensional building layout to generate a three-dimensional building layout as an output three-dimensional building model.
A method for automatically generating a three-dimensional house layout according to claim 1, wherein the architectural structural requirements further include style requirements;

The three-dimensional rendering according to the two-dimensional building layout is performed to generate a three-dimensional building layout, and the output three-dimensional building model includes:

Obtain style requirements, and convert the style requirements into decoration texture pictures for the floors and walls of each room;

Rendering the two-dimensional building layout by using the decoration texture image, and generating a three-dimensional model according to a preset height;

Retrieving a three-dimensional furniture model from a preset home database according to the style requirements;

The three-dimensional furniture model is mapped to the three-dimensional model to obtain a three-dimensional architectural layout as an output three-dimensional architectural model.
A method for automatically generating a three-dimensional house layout according to claim 2, wherein said mapping said three-dimensional furniture model to said three-dimensional model comprises:

mapping the 3D furniture model to the 3D model according to a 3D transformation matrix;

Among them, the expression of mapping is as follows:

In the formula,
Represents the 3D furniture model, which is composed of 3D coordinates; q i represents the position of the 3D furniture model in the 3D model; T represents the 3D transformation matrix.
A method for automatically generating a three-dimensional house layout according to claim 1, wherein the layout requirements include room types and room numbers for the room layout, and the outline requirements include the geometric outline of a given site;

The analyzing the layout requirements and the outline requirements based on the building data set, and searching for building bubble diagrams that meet the layout requirements and the outline requirements include:

Search in the building data set according to the layout requirements, and obtain the first building bubble map and outline pair data that meet the user's room type and room number;

According to the profile requirements, the first building bubble map and the profile data are sorted to obtain the building bubble map data matching the profile requirements;

The building bubble map data is further adjusted according to the outline requirement to obtain a building bubble map meeting the layout requirement and the outline requirement.
A method for automatically generating a three-dimensional house layout according to claim 1, wherein said obtaining a two-dimensional building layout according to the obtained building bubble diagram and said outline requirements includes:

Using a pre-trained two-dimensional layout feature generation model to perform feature extraction on the building bubble map and the outline requirements to obtain the bounding box of the room layout;

The bounding box of the room layout is fused according to the outline requirement to obtain a two-dimensional building layout.
A method for automatically generating a three-dimensional house layout according to claim 5, wherein the two-dimensional layout feature generation model is obtained in the following manner:

Use the first linear layer and the first pooling layer to construct a graph neural network;

Construct a recurrent neural network using the second linear layer, linear rectification layer, second pooling layer, and batch normalization layer;

Constructing a two-dimensional layout feature generation model according to the graph neural network and the cyclic neural network, inserting a bounding box regression loss function into the two-dimensional layout feature generation model as a target loss function;

Combining the target loss function with a training set to train the two-dimensional layout feature generation model to obtain the trained two-dimensional layout feature generation model;

Among them, the training set includes building bubble map and contour data.
A method for automatically generating a three-dimensional house layout according to claim 5, wherein the feature extraction of the building bubble map and the outline requirements by using a pre-trained two-dimensional layout feature generation model includes :

Converting the user's profile requirement into the form of a profile point set, extracting features of the profile through the constructed cyclic neural network;

Feature extraction is performed on the building bubble map through the graph neural network to obtain features of the building bubble map.
A system for automatically generating a three-dimensional house layout, characterized in that it includes:

A requirement acquisition module, configured to acquire the user's architectural structural requirements, wherein the architectural structural requirements include layout requirements and outline requirements;

A data search module, configured to analyze the layout requirements and the outline requirements based on a building dataset, and search for building bubble diagrams that meet the layout requirements and the outline requirements;

A two-dimensional layout module, configured to obtain a two-dimensional building layout according to the obtained building bubble diagram and the outline requirements;

The three-dimensional layout module is used to perform three-dimensional rendering according to the two-dimensional building layout, and generate a three-dimensional building layout as an output three-dimensional building model.
A device for automatically generating a three-dimensional house layout, characterized in that it includes:

at least one processor;

at least one memory for storing at least one program;

When the at least one program is executed by the at least one processor, the at least one processor implements the method according to any one of claims 1-7.
A computer-readable storage medium, in which a processor-executable program is stored, wherein the processor-executable program is used to execute any one of claims 1-7 when executed by a processor. method.