Decoration Style Customization Method, Its Device, Equipment, Medium, and Product

Abstract

The present application discloses a decoration style customization method, its device, equipment, medium, and product. The method includes: invoking a feature extraction model to extract the user's personal feature information and the user's historical behavior information of the target user, so as to obtain a user comprehensive feature vector; invoking a decoration style classification model to obtain the decoration style classification data corresponding to the user comprehensive feature vector; according to the decoration style type with the largest similarity probability value in the decoration style classification data, querying a set of style pattern models in the style pattern model library that matches the decoration style type; according to the house feature information of the target user, obtaining a target style pattern model in the set of style pattern models that matches the house feature information, and then generating corresponding decoration style pattern information. The present application intelligently customizes the corresponding decoration style for users based on neural network technology, and provides corresponding three-dimensional models and renderings for users to browse, improving the user experience of users on the platform.

Description

Technical Field

[0001] The present application relates to the technical field of online e-commerce, and in particular to a method for customizing decoration styles. In addition, the present application also relates to an apparatus, a device, a non-volatile storage medium, and a computer program product corresponding to the method. Background Art

[0002] With the development of Internet technology, there are many online decoration platforms on the Internet to serve Internet users who need to decorate their houses. Most online decoration platforms provide online decoration customization services for platform users so that users can customize corresponding decoration plans. However, the existing online decoration customization services provided by online decoration platforms usually only allow users to select from the decoration style types available on the platform, and then push the decoration effect pictures or three-dimensional models of the decoration styles selected by the users. Such a simple decoration customization service cannot be associated with the actual decoration needs of users, and users cannot customize an ideal decoration plan in this decoration customization service.

[0003] In view of the problems existing in the decoration customization service in the existing online decoration platforms, the applicant has made corresponding explorations in consideration of solving this problem. Summary of the Invention

[0004] The purpose of the present application is to provide a method for customizing decoration styles to meet the needs of users. In addition, the present application also relates to an apparatus, a device, a non-volatile storage medium, and a computer program product corresponding to the method.

[0005] To achieve the purpose of the present application, the following technical solutions are adopted:

[0006] A method for customizing decoration styles proposed to meet the purpose of the present application includes the following steps executed by a server:

[0007] Call a feature extraction model that has been trained to convergence to extract the user's personal feature information and user historical behavior information of the target user, so as to obtain the user comprehensive feature vector of the target user;

[0008] Call a decoration style classification model that has been trained to convergence to obtain the decoration style classification data corresponding to the user comprehensive feature vector, and the decoration style classification data includes the similarity probability values between the user comprehensive feature vector and each decoration style type;

[0009] According to the decoration style type with the largest similarity probability value in the decoration style classification data, query a set of style model sets in the style model library that match the decoration style type, and multiple style models in the style model library are stored corresponding to their style identifiers;

[0010] According to the housing feature information of the target user, obtain a target style model that matches the housing feature information from the style model set, and then generate corresponding decoration style information based on the target style model.

[0011] In a further embodiment, the method includes the following post-processing steps:

[0012] Generate a decoration style notification containing the decoration style information and push it to the target user terminal;

[0013] Respond to the notification acceptance instruction of the target user terminal acting on the decoration style notification, and generate a corresponding decoration reservation order based on the decoration style information and push it to the target user terminal;

[0014] Respond to the reservation instruction of the target user terminal acting on the decoration reservation order, and store the user order information of the target user terminal and the decoration style information in the decoration reservation database in a corresponding manner.

[0015] In a further embodiment, the method includes the following post-processing steps:

[0016] Receive the decoration style notification pushed by the server, and obtain the style model and style effect diagram included in the decoration style information in the decoration style notification;

[0017] Display a decoration style notification page in the graphical user interface to output the style model and style effect diagram to the decoration style notification page for display;

[0018] Respond to the acceptance event acting on the decoration style notification page, and push a notification acceptance instruction acting on the decoration style notification to the server.

[0019] In a further embodiment, in the step of calling the trained-to-converge decoration style classification model to obtain the decoration style classification data corresponding to the user comprehensive feature vector, the following steps performed by the decoration style classification model are included:

[0020] Receive the user comprehensive feature vector input by the server, and map the user comprehensive feature vector to the decoration style classification space through a fully connected layer;

[0021] Calculate the similarity probability values of each decoration style type in the decoration style classification space through a decoration style classifier;

[0022] Extract each of the similarity probability values to generate the decoration style classification data containing these similarity probability values.

[0023] In a further embodiment, in the step of querying the style pattern model set in the style pattern model library that matches the decoration style type with the largest similarity probability value in the decoration style classification data, the following steps are executed by the server:

[0024] Parse and obtain the decoration style type with the largest similarity probability value in the decoration style classification data, and determine the target style identifier corresponding to this decoration style type;

[0025] Query one or more of the style pattern models in the style pattern model library that correspond to the target style identifier;

[0026] Generate a style pattern model set containing the style pattern models.

[0027] In a further embodiment, in the step of obtaining the target style pattern model that matches the house feature information in the style pattern model set according to the house feature information of the target user, and then generating the corresponding decoration style pattern information according to the target style pattern model, the following steps are executed by the server:

[0028] Parse and obtain the house size data and house type data included in the house feature information;

[0029] Query the target style pattern model in the style pattern model set that corresponds to the house size data and house type data;

[0030] Obtain the style pattern effect diagram corresponding to the target style pattern model, and generate decoration style pattern information including the style pattern effect diagrams of the target style pattern model set.

[0031] A decoration style customization device proposed for the purpose of adapting to the present application includes:

[0032] A comprehensive feature extraction module, configured to call a feature extraction model that has been trained to convergence to extract the user personal feature information and user historical behavior information of the target user, so as to obtain the user comprehensive feature vector of the target user;

[0033] A classification data acquisition module, configured to call a decoration style classification model that has been trained to convergence to obtain the decoration style classification data corresponding to the user comprehensive feature vector, and the decoration style classification data includes the similarity probability values between the user comprehensive feature vector and each decoration style type;

[0034] A style model acquisition module, configured to query the style pattern model set in the style pattern model library that matches the decoration style type with the largest similarity probability value in the decoration style classification data, and multiple style pattern models in the style pattern model library are stored corresponding to their style identifiers;

[0035] A style information generation module, configured to obtain a target style model matching the housing feature information of the target user from the style model set according to the housing feature information of the target user, and further generate corresponding decoration style information according to the target style model.

[0036] In a further embodiment, the classification data acquisition module includes:

[0037] A vector mapping sub-module, configured to receive the user comprehensive feature vector input by the server, and map the user comprehensive feature vector into the decoration style classification space through a fully connected layer;

[0038] A similarity probability calculation sub-module, configured to calculate the similarity probability values of each decoration style type in the decoration style classification space through a decoration style classifier;

[0039] A classification data generation sub-module, configured to extract each of the similarity probability values to generate the decoration style classification data including these similarity probability values.

[0040] In a further embodiment, the style model acquisition module includes:

[0041] A target style identifier acquisition sub-module, configured to parse and obtain the decoration style type with the largest similarity probability value in the decoration style classification data, and determine the target style identifier corresponding to the decoration style type;

[0042] A style model matching sub-module, configured to query one or more of the style models corresponding to the target style identifier in the style model library;

[0043] A model set generation sub-module, configured to include a style model set including multiple of the style models.

[0044] In a further embodiment, the style information generation module includes:

[0045] A feature message parsing sub-module, configured to parse and obtain the housing size data and housing type data included in the housing feature information;

[0046] A target model acquisition sub-module, configured to query the target style model corresponding to the housing size data and housing type data in the style model set;

[0047] A style information generation sub-module, configured to obtain the style effect diagram corresponding to the target style model, and generate decoration style information including the style effect diagram of the target style model set.

[0048] To solve the above technical problems, an embodiment of the present application further provides a computer device, including a memory and a processor. Computer-readable instructions are stored in the memory. When the computer-readable instructions are executed by the processor, the processor executes the steps of the above-mentioned decoration style customization method.

[0049] To solve the above technical problems, an embodiment of the present application further provides a storage medium storing computer-readable instructions. When the computer-readable instructions are executed by one or more processors, one or more processors execute the steps of the above-mentioned decoration style customization method.

[0050] To solve the above technical problems, an embodiment of the present application further provides a computer program product, including a computer program and computer instructions. When the computer program and computer instructions are executed by a processor, the processor executes the steps of the above-mentioned decoration style customization method.

[0051] Compared with the prior art, the advantages of the present application are as follows:

[0052] The present application constructs an online intelligent decoration customization service for the platform. By extracting the characteristic information of users, it automatically provides users with decoration plans that meet their actual house decoration needs. Specifically, by obtaining the historical behavior information of users' online consumption or page browsing on the platform and the personal characteristic information edited by users on the platform, after extracting the historical behavior information and personal characteristic information through a feature extraction model constructed based on a neural network to generate comprehensive features, it calls a decoration style classification model also constructed based on a neural network to determine the decoration style type most consistent with the comprehensive features of users, and obtains a three-dimensional model of the target style that conforms to the decoration style type and the house characteristics of users from the style model library, so as to facilitate subsequent pushing the three-dimensional model of the target style and decoration effect pictures to users for browsing and reference to customize their decoration plans; It can be seen that compared with the existing decoration customization services, the online intelligent decoration customization service provided by the present application can automatically provide users with personalized decoration plans that match their decoration expectations. For users, the online intelligent decoration customization service provided by the platform can effectively improve the customization experience of users in the online decoration platform. For the platform, it can effectively increase the number of decoration orders in the online decoration platform. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] The above and / or additional aspects and advantages of the present application will become apparent and easy to understand from the following description of the embodiments in conjunction with the drawings, where:

[0054] Figure 1 A schematic diagram of a typical network deployment architecture related to implementing the technical solution of the present application;

[0055] Figure 2Schematic flowchart of a typical embodiment of the decoration style customization method of the present application;

[0056] Figure 3 Schematic flowchart formed by the related implementation manners of decoration style notification and decoration reservation order in the present application;

[0057] Figure 4 Schematic diagram of the graphical user interface of the decoration style notification page for displaying the effect diagram of the display style pattern in the present application;

[0058] Figure 5 Schematic diagram of the graphical user interface of the decoration style notification page for displaying the decoration style model of the display style pattern in the present application;

[0059] Figure 6 Schematic flowchart formed by the specific implementation manners of outputting and displaying the effect diagram of the display style pattern and the display style model for the target client in the present application;

[0060] Figure 7 Schematic flowchart formed by the specific implementation manners of inferring the decoration style classification data corresponding to the comprehensive feature vector of the user by the decoration style classification model in the present application;

[0061] Figure 8 Schematic flowchart formed by the specific implementation manners of the server querying from the style pattern model library to generate a set of style pattern models in the present application;

[0062] Figure 9 Schematic flowchart formed by the specific implementation manners of the server generating decoration style pattern information in the present application;

[0063] Figure 10 Principle block diagram of a typical embodiment of the decoration style customization device of the present application;

[0064] Figure 11 Basic structural block diagram of a computer device according to an embodiment of the present application. Detailed implementation manners

[0065] The embodiments of the present application will be described in detail below. The examples of the embodiments are shown in the accompanying drawings, where the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present application and should not be construed as limiting the present application.

[0066] Those skilled in the art can understand that, unless specifically stated otherwise, the singular forms "a", "an", "the" and "said" used herein may also include the plural forms. It should be further understood that the term "comprising" used in the specification of the present application means the presence of the stated features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or their groups. It should be understood that when we say that an element is "connected" or "coupled" to another element, it can be directly connected or coupled to other elements, or there may also be intermediate elements. In addition, the "connection" or "coupling" used herein may include wireless connection or wireless coupling. The phrase "and / or" used herein includes all or any unit and all combinations of one or more of the associated listed items.

[0067] Those skilled in the art can understand that, unless otherwise defined, all terms (including technical terms and scientific terms) used herein have the same meaning as the general understanding of those of ordinary skill in the art to which the present application belongs. It should also be understood that terms such as those defined in a general dictionary should be understood to have a meaning consistent with the meaning in the context of the prior art, and will not be interpreted in an idealized or overly formal sense unless specifically defined as here.

[0068] Those skilled in the art can understand that the "client", "terminal", and "terminal device" used herein include both devices with a wireless signal receiver that only has the ability to receive and no ability to transmit, and devices with both receiving and transmitting hardware that can conduct two-way communication on a two-way communication link. Such devices may include: cellular or other communication devices such as personal computers and tablet computers, which may have a single-line display or a multi-line display or a cellular or other communication device without a multi-line display; PCS (Personal Communications Service), which can combine voice, data processing, fax, and / or data communication capabilities; PDA (Personal Digital Assistant), which may include a radio frequency receiver, a pager, Internet / intranet access, a web browser, a notepad, a calendar, and / or a GPS (Global Positioning System) receiver; traditional laptop and / or palm computers or other devices, which are traditional laptop and / or palm computers or other devices with and / or including a radio frequency receiver. The "client", "terminal", and "terminal device" used herein can be portable, transportable, installed in a vehicle (air, sea, and / or land), or suitable for and / or configured to operate locally, and / or operate in a distributed manner at any other location on the earth and / or in space. The "client", "terminal", and "terminal device" used herein can also be a communication terminal, an Internet access terminal, a music / video playback terminal, such as a PDA, a MID (Mobile Internet Device), and / or a mobile phone with music / video playback functions, or can also be devices such as a smart TV and a set-top box.

[0069] The hardware referred to by names such as "server", "client", and "working node" in this application is essentially an electronic device with the equivalent capabilities of a personal computer, and is a hardware device with the necessary components disclosed by the von Neumann principle, including a central processing unit (including an arithmetic unit and a controller), a memory, an input device, and an output device. The computer program is stored in its memory, and the central processing unit loads the program stored in the external memory into the internal memory for execution, executes the instructions in the program, and interacts with the input / output devices to complete specific functions.

[0070] It should be noted that the concept of "server" referred to in this application can similarly be extended to the case applicable to a server cluster. According to the network deployment principle understood by those skilled in the art, the servers should be logically divided. Physically, these servers can either be independent of each other but can be invoked through interfaces, or be integrated into a physical computer or a set of computer clusters. Those skilled in the art should understand this flexibility and should not be restricted by this when implementing the network deployment method of this application.

[0071] Please refer to Figure 1 , the hardware foundation required for the implementation of the related technical solutions of this application can be deployed according to the architecture shown in the figure. The server 80 referred to in this application is deployed in the cloud. As a business server, it can be responsible for further connecting relevant data servers and other servers providing relevant support, etc., so as to form a logically related service cluster to provide services for relevant terminal devices such as the smart phone 81 and personal computer 82 shown in the figure or a third-party server (not shown). The smart phone and the personal computer can both access the Internet through well-known network access methods and establish a data communication link with the server 80 in the cloud to run the terminal application programs related to the services provided by the server.

[0072] For the server, the application program is usually constructed as a service process, and corresponding program interfaces are opened for the application programs running on various terminal devices to make remote calls. The related technical solutions suitable for running on the server in this application can be implemented in the server in this way.

[0073] The application program refers to the application program running on the server or terminal device. This application program implements the related technical solutions of this application in a programming way. Its program code can be saved in a non-volatile storage medium recognizable by a computer in the form of computer-executable instructions and be called into the memory by the central processing unit to run. The related device of this application is constructed through the running of this application program on the computer.

[0074] For the server, the application program is usually constructed as a service process, and corresponding program interfaces are opened for the application programs running on various terminal devices to make remote calls. The related technical solutions suitable for running on the server in this application can be implemented in the server in this way.

[0075] Those skilled in the art should be aware that: Although the various methods of this application are described based on the same concept and thus show commonality with each other, unless otherwise specified, these methods can be executed independently. Similarly, for each embodiment disclosed in this application, they are all proposed based on the same inventive concept. Therefore, concepts with the same expression, as well as concepts that are only appropriately transformed for convenience although the concept expressions are different, should be equivalently understood.

[0076] Please refer to Figure 2 , in a typical embodiment of a decoration style customization method of this application, it includes the following steps executed by the server:

[0077] Step S11, call the feature extraction model that has been trained to convergence, extract the user personal feature information and user historical behavior information of the target user, so as to obtain the user comprehensive feature vector of the target user:

[0078] The server obtains the user personal feature information and the user historical behavior information of the target user from the user database of the platform. The user personal feature information includes user age data, user gender data, user hobby data, etc.

[0079] Correspondingly, the user historical behavior information is the historical behavior data of the target user in the platform. Specifically, the platform has online shopping services for housing decoration and household appliances for users to consume. Therefore, the user historical behavior information includes the online shopping data of the target user's consumed and browsed products in the online shopping service. In addition, the platform also has an online decoration consultation service for users to use. Therefore, the user historical behavior information also includes the consultation chat data of the target user when conducting the online decoration consultation service; of course, this platform is mainly an online platform related to housing. Those skilled in the art can collect data related to housing decoration according to the relevant services provided by the platform to design the data included in the user historical behavior information, which will not be elaborated here.

[0080] After the server obtains the user personal feature information and user historical behavior information of the target user, it will call the feature extraction model that has been trained to convergence. The feature extraction model is generally constructed based on a neural network. For example, image feature extraction methods such as SIFT, ORB, or HOG are applied to extract the image data included in the user personal feature information and user historical behavior information, and text feature extraction methods such as TF-IDF, N-Gram, or NLP are applied to extract the text data included in the user personal feature information and user historical behavior information.

[0081] After the feature extraction model extracts the feature vectors corresponding to the user's personal feature information and the user's historical behavior information respectively, the feature vectors of both parties are fused to generate the user's comprehensive feature vector that can comprehensively represent the features of the user's personal feature information and the user's historical behavior information. When the feature extraction model fuses the feature vectors of both parties, it generally uses feature fusion algorithms such as TFN or LMF for feature fusion.

[0082] Step S12, call the trained-to-converge decoration style classification model to obtain the decoration style classification data corresponding to the user's comprehensive feature vector. The decoration style classification data includes the similarity probability values between the user's comprehensive feature vector and each decoration style type:

[0083] After the server obtains the user's comprehensive feature vector, it will call the trained-to-converge decoration style classification model to determine the similarity probability values between the user's comprehensive feature vector and each decoration style type existing in the platform.

[0084] The decoration style classification model is generally a model constructed based on a neural network. Specifically, the decoration style classification model contains one or more classifiers to calculate the similarity probability values between the user's comprehensive feature vector and each decoration style type. For example, the classifier is constructed using neural networks such as CNN, RNN, or FCNN, covering the technical field of machine learning.

[0085] The decoration style types generally include decoration styles such as Chinese style type, European style type, minimalist style type, and Mediterranean style type, etc. Those skilled in the art can expand the decoration style types according to the existing decoration styles, and can expand from the overall decoration style to sub-decoration styles. For example, the sub-decoration style types that can be expanded under the Chinese style type are modern Chinese style and new Chinese classical style, etc.

[0086] Regarding the decoration style classification model calculating the similarity probability values between the user's comprehensive feature vector and each decoration style type, specifically, after the server calls the decoration style classification model, it inputs the user's comprehensive feature vector into the decoration style classification model to drive the decoration style type model to fully connect the user's comprehensive feature vector and map it into the decoration style classification space. The decoration style classifier associated with the decoration style type control calculates the similarity probability values corresponding to each decoration style type of the user's comprehensive feature vector in the decoration style classification space, and then extracts these similarity probability values to generate the decoration style classification data containing these similarity probability values.

[0087] The decoration style classification data includes the similarity probability values between the user comprehensive feature vector and each decoration style type. Specifically, the sum of the similarity probability values of all decoration style types included in the decoration style classification data is generally 1, and the similarity probability value and the style identifier corresponding to its corresponding decoration style type are stored correspondingly in the decoration style classification data, so as to query the corresponding style model in the style model library through the style identifier included in the decoration style classification data later.

[0088] Step S13, according to the decoration style type with the largest similarity probability value in the decoration style classification data, query the style model set in the style model library that matches this decoration style type. Multiple style models in the style model library are stored corresponding to their style identifiers:

[0089] After the server obtains the decoration style classification data, it will determine the decoration style type with the largest similarity probability value in the decoration style classification data, obtain the style identifier of this decoration style type, so as to query one or more style models corresponding to this style identifier in the style model library, and generate the style model set including these style models.

[0090] Multiple mapping relation data composed of style identifiers and style models are stored in the style model library. The style identifier is used to represent the decoration style type of the style model stored correspondingly to it, so that the server can query the style models with this style identifier from the style model library according to the style identifier corresponding to the decoration style type with the largest similarity probability value in the decoration style classification data to form the style model set.

[0091] The style model is a three-dimensional model designed and generated by designers through modeling software. The style model is a three-dimensional model of a house with corresponding house size and house type, and furniture models and soft and hard decoration models corresponding to its decoration style type are included in this house three-dimensional model. For example, in the style model with the decoration style type of Chinese style, there will be models that conform to the Chinese style type such as mahogany furniture models and mahogany-style soft and hard decoration models; and the style model has house feature data such as its house size and house type, and these house feature data are different from the decoration style type.

[0092] Multiple style models in the style model library are classified and stored. By storing the style model corresponding to the style identifier representing its decoration style type, the style models stored in the library are classified by style.

[0093] The specific way for the server to obtain the style model set is as follows: The server parses and obtains the decoration style type with the largest similarity probability value in the decoration style classification data, determines the target style identifier corresponding to this decoration style type, and queries one or more of the style models corresponding to the target style identifier from the style model library, so as to generate a style model set containing the style models.

[0094] Step S14, according to the housing feature information of the target user, obtain a target style model that matches the housing feature information from the style model set, and then generate corresponding decoration style information according to the target style model:

[0095] After the server obtains the style model set, it will obtain the housing feature information customized and edited by the target user. The housing feature information includes housing feature data such as the housing size data, housing type data, housing floor height data, and housing geographical location data corresponding to the housing to be decorated by the target user.

[0096] After the server obtains the housing feature information, it will obtain a target style model that matches the housing feature information from the housing style model set. Specifically, the server parses and obtains the housing feature data such as the housing size data and housing type data included in the housing feature information, and queries one or more style models in the housing style model set that have the housing features characterized by these housing feature data as the target style model. For example, when the housing size data is 90 square meters and the housing type data is two living rooms and two bedrooms, the target style model with these housing feature data will be obtained from the housing style model set; of course, the server can further improve the matching accuracy of the target style model to make the matched target style model more compatible with the housing feature information. For example, when the housing size data is 85 square meters, the housing type data is two living rooms, two bedrooms, one bathroom, and one kitchen, and the housing type data also records that one living room is 10 square meters, the second living room is 20 square meters, one bedroom is 15 square meters, the second bedroom is 20 square meters, the bathroom is 5 square meters, and the kitchen is 8 square meters. At the same time, the server also refers to the housing floor height data and housing geographical location to match one or more target style models that are closest to these housing feature data from the housing style model set.

[0097] After the server obtains the target style model, it will obtain the style effect diagram corresponding to the target style model. The style effect diagram shows the target style model in the form of a picture. Regarding the acquisition method of the style effect diagram, it can generally be queried and obtained from the effect diagram library using the model ID of the target style model, or obtained by taking screenshots from the corresponding perspective angles of the 3D model of the target style model.

[0098] After the server obtains the target style model and its corresponding style effect diagram, it will generate decoration style information containing the target style model and the style effect diagram, and then push the decoration style information to the client where the target user is located in the form of platform notifications, text messages, or emails, so that the client can output the target style model and the style effect diagram to the graphical user interface for display for the target user to browse.

[0099] As can be seen from this typical embodiment, this method constructs an online intelligent decoration customization service for the platform. By extracting the user's characteristic information, it automatically provides a decoration plan that meets the actual housing decoration needs of the user. Specifically, by obtaining the historical behavior information of the user's online consumption or page browsing on the platform and the personal characteristic information edited by the user on the platform, after extracting the historical behavior information and personal characteristic information through a feature extraction model constructed based on a neural network to generate comprehensive features, it calls a decoration style classification model also constructed based on a neural network to determine the decoration style type that best matches the user's comprehensive features, and obtains a target style 3D model of this decoration style type that meets the user's housing characteristics from the style model library, so as to subsequently push the target style 3D model and decoration effect diagram to the user for browsing and reference to customize their decoration plan; thus, it can be seen that the online intelligent decoration customization service provided by this method can automatically provide a decoration plan that matches the user's decoration expectations. For the user, the online intelligent decoration customization service provided by the platform can effectively improve their customization experience in the online decoration platform. For the platform, it can effectively increase the number of decoration orders in the online decoration platform.

[0100] The above typical embodiments and their variant embodiments fully disclose the implementation scheme of the decoration style customization method of this application. However, various variant embodiments of this method can still be deduced by transforming and amplifying some technical means. The following briefly describes other embodiments:

[0101] In one embodiment, please refer to Figure 3 , the method includes the following subsequent steps:

[0102] Step S08, generate a decoration style notification containing the decoration style information and push it to the target user terminal:

[0103] After the server obtains the decoration style information, it will generate a decoration style notice containing the decoration style information, and push the decoration style information to the target user terminal in the form of a platform notice for output display for the target user to browse and perform subsequent notice acceptance events.

[0104] Step S09, in response to a notice acceptance instruction of the target user terminal acting on the decoration style notice, generate a corresponding decoration reservation order according to the decoration style information and push it to the target user terminal:

[0105] After the target user terminal receives the decoration style notice, it will parse and obtain the decoration style information contained in the decoration style notice, and output the decoration style information contained in the decoration style information to the graphical user interface for display, so that the target user at the target user terminal can trigger the notice acceptance event of the decoration style notice, and the target user terminal responds to this event and pushes the notice acceptance instruction to the server.

[0106] After the server receives the notice acceptance instruction pushed by the target user terminal, it will generate a decoration reservation order containing user order information such as the house address and contact information of the target user, and the decoration reservation order will contain a decoration plan representing the decoration style information to be decorated. In addition, the decoration reservation order may contain deposit information of the amount that the target user needs to pay.

[0107] Step S10, in response to a reservation instruction of the target user terminal acting on the decoration reservation order, store the user order information of the target user terminal and the decoration style information in the decoration reservation database in a corresponding manner:

[0108] After the target user terminal receives the decoration reservation order pushed by the server, it will output the user order information, decoration plan and pricing information contained in the decoration reservation order to the graphical user interface for display, so that the target user at the target user terminal can browse and edit relevant information, and then trigger the acceptance reservation event of the decoration reservation order, so that the target user terminal responds to this event and pushes the reservation instruction to the server.

[0109] After the server responds to the reservation instruction pushed by the target user terminal, it will store the user order information of the target user terminal and the decoration style information in the decoration reservation database in a corresponding manner, so that the business personnel of the platform can obtain the user order information and the decoration style information from the decoration reservation database, and then conduct decoration business contact with the target user.

[0110] In this embodiment, after generating the decoration style information, a notification will be pushed to the user so that the user can determine whether to accept the decoration plan in the decoration style information. Subsequently, a corresponding reservation order will be pushed after the user accepts it, so as to store the user's personal information and the decoration plan correspondingly, which is convenient for business personnel to contact the user.

[0111] In one embodiment, please refer to Figure 4 , Figure 5 and Figure 6 , the method includes the following post-processing steps:

[0112] Step S08’, receive the decoration style notification pushed by the server, and obtain the style model and style effect diagram included in the decoration style information in the decoration style notification:

[0113] The target client receives the decoration style notification pushed by the server, and will parse and process the decoration style information included in the decoration style notification, and obtain the style model and style effect diagram included in the decoration style information, so as to output them to the decoration style notification page for display later.

[0114] Step S09’, display a decoration style notification page in the graphical user interface to output the style model and style effect diagram to the decoration style notification page for display:

[0115] Please refer to Figure 4 and Figure 5 , Figure 4 The page shown in is the decoration style notification page. The target user terminal outputs and displays the style effect diagram to the decoration style browsing window 401 in the decoration style notification page for display, and can use the style display switching control 402 to switch the style effect diagram displayed in the decoration style browsing window 401 to the style model displayed in the decoration style browsing window 501 as shown in Figure 5 .

[0116] Step S10’, in response to the acceptance event acting on the decoration style notification page, push a notification acceptance instruction acting on the decoration style notification to the server:

[0117] Please refer to Figure 4 and Figure 5 , when the target user at the target user terminal touches the acceptance plan control 403 in Figure 4 or the acceptance plan control 503 in Figure 5 , the target user terminal will respond to the acceptance event and push the notification acceptance instruction to the server.

[0118] In this embodiment, after the client receives the decoration style notification, it will display the style model and the effect picture of the style on the page for the user to browse, and provide relevant controls to facilitate the user to receive the decoration style notification displayed on the page.

[0119] In one embodiment, please refer to Figure 7 , in the step of calling the trained-to-converge decoration style classification model to obtain the decoration style classification data corresponding to the user comprehensive feature vector, the following steps executed by the decoration style classification model are included:

[0120] Step S121, receive the user comprehensive feature vector input by the server, and map the user comprehensive feature vector into the decoration style classification space through a fully connected layer:

[0121] After the decoration style classification model obtains the user comprehensive feature vector, it will perform a full connection through a fully connected layer and map the user comprehensive feature vector into the decoration style classification space.

[0122] Step S122, calculate the similarity probability values of each decoration style type in the decoration style classification space through a decoration style classifier:

[0123] The decoration style classification model generally calculates the similarity probability values corresponding to each decoration style type in the decoration style classification space by means of the decoration style classifier constructed based on a multi-classifier such as Softmax.

[0124] Step S123, extract each of the similarity probability values to generate the decoration style classification data including these similarity probability values:

[0125] After the decoration style classification model maps the user comprehensive feature vector into the similarity probability values corresponding to each decoration style type in the decoration style classification space, it will extract these similarity probability values to output the decoration style classification data, which constitutes the probability distribution data corresponding to the user feature vector.

[0126] In this embodiment, the decoration style classification model calculates the similarity probability values between the user comprehensive feature vector and each decoration style type, so as to match the corresponding style model set for the target user subsequently.

[0127] In one embodiment, please refer to Figure 8 , in the step of querying the style model set in the style model library that matches the decoration style type according to the decoration style type with the largest similarity probability value in the decoration style classification data, the following steps executed by the server are included:

[0128] Step S131: Parse and obtain the decoration style type with the highest similarity probability value in the decoration style classification data, and determine the target style identifier corresponding to this decoration style type:

[0129] The server parses the decoration style classification data, determines the decoration style type with the highest similarity probability value in this decoration style classification data. This decoration style type is the decoration style type that is most similar to the comprehensive characteristics of the target user, and then determines the target style identifier corresponding to this decoration style type.

[0130] Step S132: Query one or more of the style model libraries corresponding to the target style identifier in the style model library:

[0131] After the server determines the target style identifier, it will query one or more of the style models with this target style identifier in the style model library.

[0132] Step S133: Generate a style model set containing the style models:

[0133] After the server queries the style models with the target style identifier from the style model library, it will generate the style model set containing these style models, so as to subsequently match the target style model corresponding to the housing feature information of the target user from this style model set.

[0134] In this embodiment, by querying multiple style models that are most similar to the characteristics of the target user from the style model library to generate the corresponding style model set, it is then convenient to subsequently match the corresponding target style model from this style model set.

[0135] In one embodiment, please refer to Figure 9 , in the step of obtaining the target style model corresponding to the housing feature information of the target user in the style model set and then generating the corresponding decoration style information according to this target style model, the following steps are included and executed by the server:

[0136] Step S141: Parse and obtain the housing size data and housing type data included in the housing feature information:

[0137] After the server generates the style model set, it will obtain the user feature information of the target user to parse the house feature information, and obtain the house size data and house type data included in the house feature information. The house size data is generally edited by the target user to store the size data of the house to be decorated in the platform. Correspondingly, the house type data is also generally the data edited by the target user to store the house type data of the house to be decorated in the platform.

[0138] Step S142, query the target style model corresponding to the house size data and house type data in the style model set:

[0139] Since the style model has house feature data such as the house size and house type of the house representing its decoration style, the server can query the target style model corresponding to the house size data and house type data from the style model set according to the house size data and house type data.

[0140] Step S143, obtain the style effect diagram corresponding to the target style model, and generate decoration style information including the style effect diagrams of the style model set:

[0141] The way for the server to obtain the style effect diagram can generally be to query and obtain it from an effect diagram library storing mapping relationship data composed of model IDs and style effect diagrams, or to capture screenshots from the corresponding perspective angles of the 3D model of the target style model. For example, capture the style effect diagram of the decoration wall in the 3D model of the target style model, or capture the oblique view of the 3D model of the target style model as the style effect diagram, etc.

[0142] In this embodiment, by obtaining the style model that conforms to the house size and house type of the house to be decorated by the user from the style model set, information including the style model and the corresponding effect diagram is generated, and then it is pushed to the user terminal for the user to view later.

[0143] Furthermore, by functionalizing each step in the methods disclosed in the above embodiments, a decoration style customization device of the present application can be constructed. According to this idea, please refer to Figure 10, in one typical embodiment, the device includes: an integrated feature extraction module 11, configured to call a feature extraction model that has been trained to convergence to extract the user's personal feature information and the user's historical behavior information of the target user, so as to obtain the user's integrated feature vector of the target user; a classification data acquisition module 12, configured to call a decoration style classification model that has been trained to convergence to obtain the decoration style classification data corresponding to the user's integrated feature vector, where the decoration style classification data includes the similarity probability values between the user's integrated feature vector and each decoration style type; a style model acquisition module 13, configured to query, according to the decoration style type with the largest similarity probability value in the decoration style classification data, a set of style models in the style model library that matches the decoration style type, and multiple style models in the style model library are stored corresponding to their style identifiers; a style information generation module 14, configured to obtain a target style model that matches the house feature information in the set of style models according to the house feature information of the target user, and then generate corresponding decoration style information according to the target style model.

[0144] In one embodiment, the classification data acquisition module 12 includes: a vector mapping sub-module, configured to receive the user's integrated feature vector input by the server and map the user's integrated feature vector into the decoration style classification space through a fully connected layer; a similarity probability calculation sub-module, configured to calculate the similarity probability values of each decoration style type in the decoration style classification space through a decoration style classifier; a classification data generation sub-module, configured to extract each of the similarity probability values to generate the decoration style classification data including these similarity probability values.

[0145] In one embodiment, the style model acquisition module 13 includes: a target style identifier acquisition sub-module, configured to parse and obtain the decoration style type with the largest similarity probability value in the decoration style classification data, and determine the target style identifier corresponding to the decoration style type; a style model matching sub-module, configured to query one or more style models in the style model library that correspond to the target style identifier; a model set generation sub-module, configured to generate a set of style models including multiple style models.

[0146] In one embodiment, the style information generation module 14 includes: a feature message parsing sub-module, configured to parse and obtain the house size data and the house type data included in the house feature information; a target model acquisition sub-module, configured to query the target style model in the set of style models that corresponds to the house size data and the house type data; a style information generation sub-module, configured to obtain the style effect diagram corresponding to the target style model and generate decoration style information including the style effect diagram of the set of target style models.

[0147] To solve the above technical problems, an embodiment of the present application further provides a computer device for running a computer program implemented according to the decoration style customization method. For details, please refer to Figure 11 , Figure 11 which is the basic structural block diagram of the computer device in this embodiment.

[0148] As Figure 11 shown, it is a schematic internal structure diagram of the computer device. The computer device includes a processor, a non-volatile storage medium, a memory, and a network interface connected through a system bus. Among them, the non-volatile storage medium of the computer device stores an operating system, a database, and computer-readable instructions. The control information sequence can be stored in the database. When the computer-readable instructions are executed by the processor, the processor can implement a decoration style customization method. The processor of the computer device is used to provide computing and control capabilities to support the operation of the entire computer device. The memory of the computer device can store computer-readable instructions. When the computer-readable instructions are executed by the processor, the processor can execute a decoration style customization method. The network interface of the computer device is used to connect and communicate with the terminal. Those skilled in the art can understand that Figure 11 the structure shown in

[0149] is only a block diagram of some structures related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine some components, or have different component arrangements.

[0150] The present application also provides a non-volatile storage medium. The decoration style customization method is written as a computer program and stored in the storage medium in the form of computer-readable instructions. When the computer-readable instructions are executed by one or more processors, it means that the program runs on the computer, thereby enabling one or more processors to execute the steps of the decoration style customization method in any of the above embodiments.

[0151] Those of ordinary skill in the art can understand that all or part of the processes of implementing the methods in the above embodiments can be completed by instructing relevant hardware through a computer program. This computer program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above methods. Among them, the aforementioned storage medium can be a non-volatile storage medium such as a magnetic disk, an optical disc, a read-only memory (ROM), or a random access memory (RAM), etc.

[0152] In summary, this application intelligently customizes corresponding decoration styles for users based on neural network technology, and provides corresponding 3D models and renderings for users to view, improving the user experience on the platform.

[0153] It should be understood that although the steps in the flowchart of the accompanying drawings are shown in sequence according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless there is a clear indication in this article, the execution of these steps has no strict order restriction, and they can be executed in other orders. Moreover, at least some of the steps in the flowchart of the accompanying drawings may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. Their execution order is not necessarily sequential, but can be executed alternately or alternately with at least a part of other steps or sub-steps or stages of other steps.

[0154] Those skilled in the art of this technology can understand that the steps, measures, and solutions in the various operations, methods, and processes discussed in this application can be alternated, changed, combined, or deleted. Further, other steps, measures, and solutions in the various operations, methods, and processes discussed in this application can also be alternated, changed, rearranged, decomposed, combined, or deleted. Further, the steps, measures, and solutions in the prior art that are the same as those disclosed in this application can also be alternated, changed, rearranged, decomposed, combined, or deleted.

[0155] The above are only some implementation manners of this application. It should be noted that for those of ordinary skill in the art of this technology, without departing from the principle of this application, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of this application.

Claims

1. A method for customizing a decoration style, characterized in that, it includes the following steps executed by the server: Call the feature extraction model that has been trained to convergence, extract the user's personal feature information and user historical behavior information of the target user, so as to obtain the user's comprehensive feature vector of the target user; Call the decoration style classification model that has been trained to convergence, obtain the decoration style classification data corresponding to the user's comprehensive feature vector, and the decoration style classification data includes the similarity probability values between the user's comprehensive feature vector and each decoration style type; According to the decoration style type with the largest similarity probability value in the decoration style classification data, query the style style model set in the style style model library that matches this decoration style type, including the following steps executed by the server: Parse and obtain the decoration style type with the largest similarity probability value in the decoration style classification data, and determine the target style identifier corresponding to this decoration style type; Query one or more of the style style models in the style style model library that correspond to the target style identifier; Generate a style style model set including the style style models, and multiple style style models in the style style model library are stored corresponding to their style identifiers; According to the house feature information of the target user, obtain the target style style model in the style style model set that matches this house feature information, and then generate the corresponding decoration style information according to this target style style model, including the following steps executed by the server: Parse and obtain the house size data and house type data included in the house feature information; Query the target style style model in the style style model set that corresponds to the house size data and house type data; Obtain the style style effect diagram corresponding to the target style style model, and generate the decoration style information including the style style effect diagram of the target style style model set.

2. The method according to claim 1, characterized in that, this method includes the following post-processing steps: Generate a decoration style notification including the decoration style information and push it to the target user terminal; Respond to the notification acceptance instruction of the target user terminal acting on the decoration style notification, and generate a corresponding decoration reservation order according to the decoration style information and push it to the target user terminal; Respond to the reservation instruction of the target user terminal acting on the decoration reservation order, and store the user order information of the target user terminal and the decoration style information in a corresponding manner in the decoration reservation database.

3. The method according to claim 1, characterized in that, this method includes the following post-processing steps: Receive the decoration style notification pushed by the server, and obtain the style style model and style style effect diagram included in the decoration style information in this decoration style notification; Display a decoration style notification page on the graphical user interface to output the style style model and style style effect diagram to the display in this decoration style notification page; Respond to the acceptance event acting on the decoration style notification page, and push a notification acceptance instruction acting on the decoration style notification to the server.

4. The method according to claim 1, characterized in that, In the step of calling the trained-to-converge decoration style classification model to obtain the decoration style classification data corresponding to the user comprehensive feature vector, the following steps executed by the decoration style classification model are included: Receive the user comprehensive feature vector input by the server, and map the user comprehensive feature vector into the decoration style classification space through a fully connected layer; Calculate the similarity probability values of each decoration style type in the decoration style classification space through a decoration style classifier; Extract each of the similarity probability values to generate the decoration style classification data including these similarity probability values.

5. A decoration style customization device Characterized in that It includes: A comprehensive feature extraction module, configured to call a trained-to-converge feature extraction model to extract the user personal feature information and user historical behavior information of a target user, so as to obtain the user comprehensive feature vector of the target user; A classification data acquisition module, configured to call a trained-to-converge decoration style classification model to obtain the decoration style classification data corresponding to the user comprehensive feature vector, where the decoration style classification data includes the similarity probability values between the user comprehensive feature vector and each decoration style type; A style model acquisition module, configured to query, according to the decoration style type with the largest similarity probability value in the decoration style classification data, a set of style models matching the decoration style type in a style model library, including the following steps executed by the server: Parse and obtain the decoration style type with the largest similarity probability value in the decoration style classification data, and determine the target style identifier corresponding to the decoration style type; Query one or more of the style models corresponding to the target style identifier in the style model library; Generate a set of style models including the style models, and multiple style models in the style model library are stored corresponding to their style identifiers; A style information generation module, configured to obtain, according to the house feature information of the target user, a target style model matching the house feature information in the set of style models, and then generate corresponding decoration style information according to the target style model, including the following steps executed by the server: Parse and obtain the house size data and house type data included in the house feature information; Query the target style model corresponding to the house size data and house type data in the set of style models; Obtain the style effect diagram corresponding to the target style model, and generate decoration style information including the style effect diagrams of the target style model set; 6. An electronic device, including a central processing unit and a memory Characterized in that The central processing unit is configured to call and run a computer program stored in the memory to execute the steps of the method according to any one of claims 1 to 4.

7. A non-volatile storage medium Characterized in that It stores, in the form of computer-readable instructions, a computer program implemented according to the method according to any one of claims 1 to 4. When the computer program is called and run by a computer, it executes the steps included in the method.

8. A computer program product comprising a computer program / instructions, wherein, when the computer program / instructions are executed by a processor, the steps of the method according to any one of claims 1 to 4 are implemented.

Images