Promotional information display method and apparatus, computer device, and storage medium

By generating rendered images by displaying the descriptive dimensions and options of the search object, the problem of users having to adjust keywords multiple times in traditional technologies is solved, achieving accurate matching of visual decision-making assistance and promotional information, and reducing resource waste.

WO2026232977A1PCT designated stage Publication Date: 2026-11-12TENCENT TECHNOLOGY (SHENZHEN) CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
PCT/CN2026/083609
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2025-05-09
Filing Date
2026-03-16
Publication Date
2026-11-12

AI Technical Summary

Technical Problem

In traditional technology, users need to constantly adjust keywords when searching for products to see if the promotional information meets their expectations, resulting in a large number of invalid promotional messages being displayed and causing a waste of resources.

Method used

This paper provides a method for displaying promotional information. By displaying the description dimensions and options of the search object, a rendered image matching the selected description option is generated, and the matching promotional information is displayed, thereby achieving visual decision assistance and helping users transform abstract needs into concrete visual expressions.

Benefits of technology

Reduce users' cognitive load, shorten decision-making time, improve the relevance of promotional information, reduce resource waste, and enhance promotional effectiveness.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN2026083609_12112026_PF_FP_ABST
    Figure CN2026083609_12112026_PF_FP_ABST
Patent Text Reader

Abstract

A promotional information display method, performed by a computer device, the method comprising: in response to a search operation, displaying a descriptive option of at least one descriptive dimension of a search object of the search operation (202); in response to a selection operation on the descriptive option of the at least one descriptive dimension, displaying a rendered image generated according to the selected descriptive option of the at least one descriptive dimension, the rendered image comprising a virtual object matching the selected descriptive option (204); and displaying promotional information, wherein a a promotional object described by the promotional information matches the virtual object (206).
Need to check novelty before this filing date? Find Prior Art

Description

Promote information display methods, devices, computer equipment and storage media

[0001] Related applications

[0002] This application claims priority to Chinese patent application filed on May 9, 2025, with application number 202510602143.8, entitled "Method, apparatus, computer device and storage medium for displaying promotional information", the entire contents of which are incorporated herein by reference. Technical Field

[0003] This application relates to the field of computer technology, and in particular to the promotion of information methods, apparatus, computer equipment, storage media and computer program products. Background Technology

[0004] With the development of computer and internet technology, promoting products online has become an important means for businesses to promote their products and services. For example, when a user searches for an item using a computer device, the device can display promotional information based on the user's search intent. For instance, when a user searches for athletic shoes, the device can display information about athletic shoe products.

[0005] In traditional technologies, users often don't know exactly what they want to buy. They often have to constantly adjust the keywords they enter and search multiple times to see if the displayed promotional information meets their expectations. This results in a large number of invalid promotional messages being displayed, leading to a waste of promotional resources. Summary of the Invention

[0006] Therefore, it is necessary to provide promotional information display methods, devices, computer equipment, computer-readable storage media, and computer program products that can reduce the waste of promotional resources in response to the above-mentioned technical problems.

[0007] Firstly, this application provides a method for displaying promotional information. The method includes:

[0008] In response to a search operation, display description options for at least one description dimension of the search object for the search operation;

[0009] In response to a selection operation of a description option under the at least one description dimension, a rendered image generated according to the selected description option under the at least one description dimension is displayed, the rendered image containing a virtual object matching the selected description option;

[0010] The promotional information is displayed, and the promotional object described in the promotional information matches the virtual object.

[0011] Secondly, this application also provides a promotional information display device. The device includes:

[0012] A description option display module is used to display description options for at least one description dimension of the search object in response to a search operation;

[0013] A rendering image display module is configured to, in response to a selection operation of a description option under the at least one description dimension, display a rendered image generated according to the selected description option under the at least one description dimension, the rendered image containing a virtual object matching the selected description option;

[0014] The promotion information display module is used to display promotion information, wherein the promotion object described in the promotion information matches the virtual object.

[0015] Thirdly, this application also provides a computer device. The computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps of the above-described promotional information display method.

[0016] Fourthly, this application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program thereon, which, when executed by a processor, implements the steps of the above-described promotional information display method.

[0017] Fifthly, this application also provides a computer program product. The computer program product includes a computer program that, when executed by a processor, implements the steps of the above-described promotional information display method.

[0018] The aforementioned promotional information display method, apparatus, computer equipment, storage medium, and computer program product, in response to a search operation, displays description options for at least one descriptive dimension of the search object, thereby breaking down the user's complex decision-making process into operable, simplified steps, reducing the user's cognitive load. Furthermore, in response to a selection operation of a description option under at least one descriptive dimension, it displays a rendered image generated according to the selected description option. The rendered image contains virtual objects matching the selected description option. Thus, the instant feedback from the rendered image provides visual decision support, helping users transform abstract needs into concrete visual expressions, shortening the cognitive loop from intent formation to decision confirmation. The promotional information, whose description of the promotional object matches the virtual objects in the rendered image, effectively matches the user's search intent, thereby significantly improving the promotional effect of the information and reducing the waste of promotional resources.

[0019] Sixthly, this application also provides another method for displaying promotional information, the method comprising:

[0020] Display promotional information, and description options for at least one description dimension of the promotional object corresponding to the promotional information;

[0021] In response to a selection operation of a description option under the at least one description dimension, a rendered image generated according to the selected description option under the at least one description dimension is displayed;

[0022] Based on the rendered image, match at least one target promotion object and display it.

[0023] Seventhly, this application also provides another promotional information display device, the device comprising:

[0024] The promotion information display module is used to display promotion information and description options for at least one description dimension of the promotion object corresponding to the promotion information;

[0025] A rendering image display module is configured to display a rendered image generated according to the selected description option under the at least one description dimension in response to a selection operation of the description option under the at least one description dimension.

[0026] The promotion target matching module is used to match at least one target promotion target based on the rendered image and display it.

[0027] Eighthly, this application also provides a computer device. The computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps of the above-described promotional information display method.

[0028] Ninthly, this application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program thereon, which, when executed by a processor, implements the steps of the above-described promotional information display method.

[0029] Tenthly, this application also provides a computer program product. The computer program product includes a computer program that, when executed by a processor, implements the steps of the above-described promotional information display method.

[0030] The aforementioned promotional information display method, apparatus, computer equipment, storage medium, and computer program product, by displaying promotional information and description options for at least one description dimension of the promotional object corresponding to the promotional information, can decompose the user's complex decision-making process into operable and simplified steps, reducing the user's cognitive load. Furthermore, in response to the selection operation of description options under at least one description dimension, a rendered image generated according to the selected description option under at least one description dimension is displayed. This allows for visual decision assistance through the immediate feedback of the rendered image, helping users transform abstract needs into concrete visual expressions, shortening the cognitive loop from intention formation to decision confirmation. Furthermore, based on the rendered image, at least one target promotional object is matched and displayed, thereby enabling accurate recommendation of promotional objects that meet the user's needs based on the displayed promotional information. Attached Figure Description

[0031] To more clearly illustrate the technical solutions in the embodiments or related technologies of this application, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0032] Figure 1 is an application environment diagram of the promotion information display method in some embodiments;

[0033] Figure 2 is a flowchart illustrating the promotional information display method in some embodiments;

[0034] Figure 3 is a schematic diagram of the interface displaying description options in some embodiments;

[0035] Figure 4 is a schematic diagram of the interface displaying the preview image in some embodiments;

[0036] Figure 5 is a comparison of rendered images as a schematic diagram of the interface in some embodiments;

[0037] Figure 6 is a schematic diagram of the interface displaying scene rendering images in some embodiments;

[0038] Figure 7 is a schematic diagram of the user interface for searching for products in some embodiments;

[0039] Figure 8 is a schematic diagram of the interface changes when a user makes a selection change operation in some embodiments;

[0040] Figure 9 is a schematic diagram of the interface for displaying rendered images and viewing product details in some other embodiments;

[0041] Figure 10 is a schematic diagram of the user interface for searching for products in some other embodiments;

[0042] Figure 11 is a system architecture diagram involving the promotional information display method in some embodiments;

[0043] Figure 12 is an interaction sequence diagram involving the promotion information display method in some embodiments;

[0044] Figure 13 is a flowchart illustrating the promotion information display method in some other embodiments;

[0045] Figure 14 is a structural block diagram of the promotion information display device in some other embodiments;

[0046] Figure 15 is a structural block diagram of the promotion information display device in some embodiments;

[0047] Figure 16 is an internal structure diagram of a computer device in some embodiments;

[0048] Figure 17 is an internal structure diagram of a computer device in some other embodiments. Detailed Implementation

[0049] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0050] First, a brief introduction to some terms used in the embodiments of this application will be given:

[0051] Description dimension: This refers to the dimension used to describe the search object desired by the user. In this embodiment, at least one description dimension is provided for the user to describe the object they want to search for, and multiple description options can be displayed under each description dimension. For example, at least one description dimension may include a style dimension and a color scheme dimension.

[0052] Description Options: These refer to the specific options under each description dimension. Each description dimension can display one or more description options, each describing a different feature within that dimension. For example, the style dimension can display description options that describe abstract style attributes of a product or scene, such as "forest style" or "cyberpunk."

[0053] Rendered images: refers to two-dimensional / three-dimensional visual content generated in real time through artificial intelligence technology. For example, it can generate scene-based product display images that conform to the target aesthetic characteristics based on the style and color options selected by the user, and support the fusion of light and shadow effects and the adaptation of composition proportions.

[0054] Virtual objects are objects created and existing in the virtual environment of a rendered image, and are not part of the real physical world. Specifically, a virtual object can be a product within a rendered image. For example, when a user's search query targets sneakers, the sneakers in the rendered image would be a virtual object.

[0055] Promotional information refers to marketing or promotional content proactively disseminated by businesses, organizations, or individuals to a target audience through various channels to achieve specific goals (such as enhancing brand awareness, boosting product sales, or disseminating ideas). Promotional information can specifically be product data used for advertising, such as product promotional information, product descriptions, or product links—any one or more of these. Promotional information can be in text, image, or video format. There can be one or more promotional messages, with "multiple" meaning at least two.

[0056] Target audience: This refers to the object being promoted by the promotional information. For example, when the promotional information is product data used as an advertisement, the target audience is the product being promoted. For instance, if the promotional information is an advertisement for athletic shoes, then the target audience is the athletic shoes being promoted.

[0057] Cross-modal product matching: Analyze the visual features of the rendered image and the semantic features of the description options selected by the user, establish a visual-semantic joint mapping relationship, and filter products from the product library that simultaneously meet explicit parameters (such as color and material) and implicit features (style and tone).

[0058] Real-time recommendation engine: Based on multi-dimensional parameters generated during the user customization process (such as style preferences, color scheme selection, and composition preferences), the engine dynamically adjusts the product filtering strategy and ranking weight to ensure that the recommendation results are strongly correlated with the user's real-time intent.

[0059] This application provides a promotional information display method that can be applied to the application environment shown in Figure 1. The terminal 102 and server 104 can communicate via a network, such as a wired or wireless network. The promotional information database 106 can store promotional information. The promotional information database can be set up separately, integrated into server 104, or placed in the cloud or on other servers. Those skilled in the art will understand that the number of terminals 102 can be more or less. For example, there can be only one terminal 102, or dozens or hundreds, or more. This application does not limit the number or type of terminals 102.

[0060] Terminal 102 can be, but is not limited to, various desktop computers, laptops, smartphones, tablets, IoT devices, and portable wearable devices. IoT devices can include smart speakers, smart TVs, smart air conditioners, smart in-vehicle systems, etc. Portable wearable devices can include smartwatches, smart bracelets, head-mounted devices, etc. Terminal 102 can run a target application that supports object search. This target application can be a search application specifically supporting object search, or other applications that provide object search functionality, such as communication applications, video applications, shopping applications, audio playback applications, etc. This application does not limit the specific application. Furthermore, this application does not limit the form of the target application; it can be a parent application running on an operating system, a sub-application running within a parent application (such as a mini-program), or a webpage.

[0061] Server 104 can be a standalone physical server, a server cluster or distributed system consisting of multiple physical servers, or a cloud server providing cloud computing services. Server 104 can be the backend server for the aforementioned target application, used to provide backend services for the target application.

[0062] The promotional information display method provided in this application embodiment can be executed by a computer device. A computer device refers to an electronic device with data computing, processing and storage capabilities. A computer device can be a terminal, a server, or a system composed of a terminal and a server.

[0063] Taking the application environment shown in Figure 1 as an example, terminal 102 can execute the promotion information display method of this application with the cooperation of server 104. The terminal responds to the search operation through the target application that supports object search and displays description options for at least one description dimension of the search object. In response to the selection operation of the description option under at least one description dimension, the terminal displays a rendered image generated according to the selected description option under at least one description dimension. The rendered image contains a virtual object that matches the selected description option. The rendered image can be rendered by the terminal or by the server. The server further queries the promotion information database based on the rendered image and sends the promotion information that matches the described promotion object and the virtual object to the terminal. The terminal displays the promotion information through the target application.

[0064] In some embodiments, as shown in FIG2, a method for displaying promotional information is provided. This method is executed by a computer device, which may be a server 104 or a terminal 102 in FIG1. ​​In this embodiment, the method is described using the terminal 102 in FIG1 as an example, and includes the following steps:

[0065] Step 202: In response to the search operation, display description options for at least one description dimension of the search object of the search operation.

[0066] The search operation is a series of actions and processes for finding, locating, and retrieving specific objects. In practice, the search operation can be an action by which a user provides search conditions or keywords through input boxes, voice input, etc. The search object refers to the target or thing targeted by the search operation, which can be various types of entities, such as various types of goods. At least one descriptive dimension can be one or more descriptive dimensions. A descriptive dimension refers to the dimension that describes the search object the user wants. In this embodiment, at least one descriptive dimension is provided for the user to describe the object they want to search for, and multiple descriptive options can be displayed under each descriptive dimension. For example, at least one descriptive dimension can include a style dimension and a color scheme dimension. Descriptive options refer to the specific options under a descriptive dimension. One or more descriptive options can be displayed under each descriptive dimension, and each descriptive option describes different features under that descriptive dimension.

[0067] Specifically, the terminal can run a target application that supports object search, and users can perform search operations through the search page displayed by the target application. For example, the search page of the target application can display a search box, where users can enter keywords to perform a search.

[0068] Step 204: In response to the selection operation of a description option under at least one description dimension, display a rendered image generated according to the selected description option under at least one description dimension, the rendered image containing virtual objects that match the selected description option.

[0069] The selection operation for description options under at least one description dimension can be either selecting description options under all description dimensions in the at least one description dimension, or selecting description options under a subset of description dimensions.

[0070] Rendered images refer to two-dimensional / three-dimensional visual content generated in real time using artificial intelligence technology. For example, based on the user's selected style and color scheme options, it can generate scene-based product display images that conform to the target aesthetic characteristics, supporting the integration of lighting and shadow effects and the adaptation of composition proportions. Rendered images are generated according to the selected description options under at least one description dimension and contain virtual objects that match the selected description options, which can visually represent the search intent indicated by the user's search operation.

[0071] A virtual object is an object that exists outside the real physical world and is created and exists within the virtual environment of a rendered image. Specifically, a virtual object can be a product within a rendered image. For example, when a user's search query targets sneakers, the sneakers in the rendered image would be a virtual object.

[0072] Specifically, after a user selects a description option under at least one description dimension, the terminal can respond to the selection operation by displaying a rendered image generated according to the selected description option under at least one description dimension, the rendered image containing a virtual object matching the selected description option.

[0073] In practical applications, the rendered image can be generated by the terminal or by the server. For example, the terminal can install a trained image rendering model, and the terminal can use the image rendering model to render the image according to the selected description options under each description dimension to obtain the rendered image. Taking a description dimension that includes at least one style dimension and one color dimension as an example, the terminal can use a cross-modal model to associate the style description text in the description options under the style dimension with the corresponding visual elements in the image feature library, thereby parsing the style description text into a specific combination of visual features. For example, "cyberpunk" can be parsed as "cyberpunk → neon lights, metallic texture, dystopian architecture". The description options under the color dimension can be converted into specific color space parameters. The HSL / HSV color model is used to define the hue, saturation, and lightness. For example, "Morandi color scheme" corresponds to a color scheme with low saturation (S = 20-40%) and medium lightness (L = 50-70%). Further, prompt words are constructed based on the description options under the style dimension and the description options under the color dimension. The prompt words are input into a large model such as Stable Diffusion. The parsed combination of visual features and color space parameters are passed to the large model as additional input to generate an image, thereby obtaining a rendered image.

[0074] It is understood that, in other embodiments, in order to save the terminal's computer resources, the process of generating the rendered image can be performed by the server. The terminal can send the selected description options under each description dimension to the server and request the server to perform image rendering to generate an image. After generating the rendered image, the server returns the rendered image to the terminal, so that the terminal can display the rendered image.

[0075] Step 206: Display promotional information, matching the promotional target described in the promotional information with the virtual object.

[0076] In this context, "promotional information" refers to marketing or promotional content proactively disseminated by businesses, organizations, or individuals to a target audience through various channels to achieve specific goals (such as enhancing brand awareness, boosting product sales, or disseminating ideas). Promotional information can specifically be product data used for advertising, such as product promotional information, product descriptions, or product links—any one or more of these. Promotional information can be in text, image, or video format. There can be one or more promotional messages, with "multiple" meaning at least two. The "promotional target" refers to the object promoted by the promotional information. For example, when the promotional information is product data used for advertising, the promotional target is the product being advertised. For instance, if the promotional information is an advertisement for athletic shoes, then the promotional target is the athletic shoes being advertised. The promotional target described in the promotional information matches the virtual object; that is, the promotional target and the virtual object are of the same type and have the same visual characteristics. For example, if the virtual object is athletic shoes of type X, style Y, and color Z, then the promotional target is also athletic shoes of type X, style Y, and color Z.

[0077] Specifically, the server can query the promotion information database based on the rendered image, and send the promotion information that matches the described promotional object with the virtual object to the terminal. The terminal can then display the promotion information through the target application.

[0078] For example, when the search object is a product, the server can perform cross-modal product matching based on the rendered image and the selected description options under each description dimension, thereby retrieving promotional information for a specific product from the product library and sending it to the terminal. Cross-modal product matching refers to parsing the visual features of the rendered image and the semantic features of the user-selected description options, establishing a visual-semantic joint mapping relationship, and filtering products from the product library that simultaneously satisfy explicit parameters (such as color and material) and implicit features (style and tone).

[0079] For example, referring to Figure 3, assuming the user's search operation is for sneakers, the terminal can respond to the search operation by displaying description options under multiple description dimensions, such as shoe type option 302 under shoe type dimension, style option 304 under style dimension, and color scheme option 306 under color scheme dimension. When the user selects a description option under each description dimension, a rendered image generated according to the selected description option under each description dimension can be displayed. For example, when the user selects "shoe type 1" under shoe type dimension, "style 2" under style dimension, and color scheme 1 under color scheme dimension, a rendered image 308 can be displayed. Furthermore, promotional information can be displayed, and the promotional information describes a promotional object that matches the sneakers in the rendered image, such as promotional information 310 in Figure 3.

[0080] It is understood that Figure 3 is merely an illustrative example of this application and is not intended to limit the implementation of this application. In some other embodiments, the terminal may sequentially display description options under each dimension. Specifically, when a user selects a description option under a certain description dimension, the terminal displays the description option associated with that description option in the next description dimension. For example, after a user selects shoe type 1 under the shoe type dimension, the terminal displays the style option associated with shoe type 1 under the style dimension. When the user continues to select a style option among the style options associated with shoe type 1, the terminal continues to display the color scheme option associated with the selected style option under the color scheme dimension, and so on. When the user completes the selection operation under all description dimensions, the terminal displays the rendered image generated according to the description options selected under each description dimension.

[0081] In the aforementioned method for displaying promotional information, since it can respond to a search operation by displaying description options for at least one descriptive dimension of the search object, the user's complex decision-making process can be broken down into actionable and simplified steps, reducing the user's cognitive load. Furthermore, in response to the selection operation of description options under at least one descriptive dimension, a rendered image generated according to the selected description option is displayed. The rendered image contains virtual objects matching the selected description option. Thus, the immediate feedback from the rendered image provides visual decision support, helping users transform abstract needs into concrete visual expressions, shortening the cognitive loop from intent formation to decision confirmation. The promotional information, which describes the promotional object and matches the virtual objects in the rendered image, effectively matches the user's search intent, thereby improving the promotional effect of the information and reducing the waste of promotional resources.

[0082] In some exemplary embodiments, in response to a search operation, displaying description options for at least one description dimension of the search object includes: in response to a search operation, displaying description options for multiple description dimensions arranged in a preset order; wherein, the description options displayed under each description dimension starting from the second description dimension are associated with the description option selected under the previous description dimension.

[0083] In this embodiment, different description dimensions can be preset for different types of search objects. For example, for the search object of footwear, three description dimensions can be preset: shoe type, style, and color scheme. The description dimensions corresponding to each type of search object are displayed in a preset order. For example, the description options of the three description dimensions corresponding to footwear are displayed in the preset order of "shoe type → style → color scheme". The description options displayed under each description dimension starting from the second description dimension are associated with the description options selected under the previous description dimension. That is, starting from the second description dimension, the optional content of each description dimension is dynamically adjusted based on the previous selection. For example, after selecting "casual" shoe type in the first dimension, the "style" option displayed in the second dimension only includes options related to casual shoe type.

[0084] In practice, after a user performs a search, the terminal can determine the search object, further determine the multiple descriptive dimensions corresponding to the search object and the preset order of the multiple descriptive dimensions, and display the descriptive options under the first descriptive dimension according to the preset order. When the user selects a descriptive option displayed under the first descriptive dimension, the descriptive options associated with the selected descriptive option under the first dimension are displayed under the second descriptive dimension. When the user selects a descriptive option under the second descriptive dimension, the descriptive options associated with the selected descriptive option under the second dimension are displayed under the third descriptive dimension, and so on, until the user completes the selection operation under the last descriptive dimension.

[0085] In the above embodiments, by guiding users through progressively increasing options, users can avoid being faced with a massive number of filtering conditions at once, reducing their cognitive load and improving the efficiency of their interaction when making selections.

[0086] In some exemplary embodiments, in response to a selection operation of a description option under at least one description dimension, displaying a rendered image generated according to the selected description option under at least one description dimension includes: in response to a selection operation of a description option under at least one description dimension, displaying description options associated with the selected description option in a set of description options under the next description dimension; in response to a selection operation of at least one description option among the associated description options, if a next description dimension exists, performing the step of displaying description options associated with the selected description option in a set of description options under the next description dimension; and if a next description dimension does not exist, displaying a rendered image generated according to each of the selected description options.

[0087] In this embodiment, the description options under each description dimension corresponding to the search object can be displayed sequentially in a preset order. Thus, the terminal can respond to the selection operation of the description option under at least one description dimension and display the description options associated with the selected description option in the set of description options under the next description dimension. The user continues to make a selection operation under the next description dimension. At this time, the terminal determines whether there are any description dimensions that have not been displayed. If there are any description dimensions that have not been displayed, the terminal continues to display the description options associated with the selected description option in the set of description options under the next description dimension. If there is no next description dimension, the terminal can display the rendered image generated according to the selected description options under each description dimension.

[0088] In the above embodiments, since only the description options associated with the currently selected description option can be displayed in the next description dimension, the interaction efficiency of the user's selection operation is improved. Only when there is no next description dimension will the rendered image generated according to each of the selected description options be displayed. The displayed rendered image can more accurately visually express the user's intention.

[0089] In some exemplary embodiments, the promotional information display method of this application further includes: in response to any description option changing from an initial state to a selected state, displaying a preview image generated according to the description option that has changed to the selected state, the preview image containing a preview object matching the description option that has changed to the selected state.

[0090] In this context, a preview image refers to a temporary image that is dynamically displayed during user interaction, providing real-time feedback on selection results or operational effects. A preview object is the specific element within the preview image that is associated with the user's action.

[0091] Specifically, when a user selects a description option under a certain description dimension, the description option can change from its initial state to a selected state. The terminal can respond to the state change of the description option by displaying a preview image generated according to the description option whose state has changed in the image display area. The preview image is generated according to the description option whose state has changed and contains preview objects that match the description option. For example, assuming that the user's search operation is for sneakers, and the description option "leopard print" under the style dimension is selected by the user and its state changes, the terminal can display sneakers with a leopard print style.

[0092] In practical applications, the description options in this embodiment can be displayed on an object search page. This object search page may also include an image display area where the terminal can display a preview image. The object search page is a page where search operations can be performed. It can display a search input box where the user can enter keywords expressing their search intent. The image display area refers to an area where images can be displayed. For example, in the object search page, the positional relationship between the image display area and the description options under each description dimension can be vertical or horizontal; this application does not limit this relationship.

[0093] For example, referring to Figure 4, the object search page 400 in Figure 4 displays an image display area 402. When the user selects the "Shoe Type 1" option under the shoe type dimension, a preview image 402 is displayed in this area. This preview image contains a preview object 406 that matches the "Shoe Type 1" option.

[0094] For example, when a user changes the selected description option under this description dimension, the preview image displayed in the image display area changes accordingly. For example, when the user selects a description option under another dimension, the preview image displayed in the image display area can be updated to the preview image corresponding to the description option under that other dimension.

[0095] For example, when any description option changes from its initial state to a selected state, and other description options in other dimensions are also in a selected state, the preview image displayed in the image display area can be a preview image generated according to the description option that has changed to a selected state and the other description options that are in a selected state. The preview object in the preview image matches the description option that has changed to a selected state and the other description options that are in a selected state.

[0096] In the above embodiments, since preview images can be displayed, real-time visual feedback can be provided to the user's selection operations, stimulating the user's awareness of unnoticed aesthetic preferences, thereby improving the interactive efficiency in the process of displaying promotional information.

[0097] In some exemplary embodiments, in response to a selection operation of a description option under at least one description dimension, displaying a rendered image generated according to the selected description option under at least one description dimension includes: in response to a selection operation of a description option under at least one description dimension, displaying the selected description option under at least one description dimension; when the selected description option under at least one description dimension satisfies a rendering trigger condition, displaying a rendering trigger control; and in response to a trigger operation of the rendering trigger control, displaying a rendered image generated according to the selected description option under at least one description dimension.

[0098] Rendering trigger conditions refer to the conditions that can trigger image rendering. These conditions can be set as needed, such as the number of selected description options reaching a preset number or the selected description options covering all description dimensions. Rendering trigger controls are interactive components that trigger image rendering, such as interactive buttons that respond to user actions. Trigger actions refer to preset actions applied to the rendering trigger controls. These preset actions can be touch actions, cursor actions, button actions, or voice actions. Touch actions can be clicks, presses, or swipes, and can be single-point or multi-point touches. Cursor actions can be clicking or pressing the cursor. Button actions can be virtual or physical buttons.

[0099] Specifically, in response to the selection of description options under at least one description dimension, the terminal can display the selected description option under at least one description dimension. When the selected description option under at least one description dimension meets the rendering trigger condition, a rendering trigger control is displayed. The user can trigger the rendering trigger control to generate a rendered image according to the selected description option under at least one description dimension. The terminal then displays the generated rendered image.

[0100] For example, the description options in this embodiment can be displayed on the object search page, which can also display an image display area. A rendering trigger control is displayed corresponding to the image display area. Since the rendering trigger control is displayed corresponding to the image display area, it can guide the user to perform a trigger operation to trigger the generation and display of the rendered image.

[0101] In the above embodiments, since a rendering trigger control can be displayed, the rendered image is only displayed after the user performs a trigger operation on the rendering trigger control, which improves the flexibility of the rendering image generation process.

[0102] In some exemplary embodiments, displaying promotional information includes: displaying a first confirmation control corresponding to the rendered image in response to the display of the rendered image; and displaying promotional information in response to a triggering operation on the first confirmation control.

[0103] The first confirmation control corresponding to the rendered image refers to an interactive component that confirms the rendered image. For example, it could be an interactive button that corresponds to the display of the rendered image and can respond to the user's trigger operation.

[0104] Specifically, after displaying the rendered image, the terminal can display a first confirmation control at a preset position around the rendered image. If the user is satisfied with the rendered image, they can trigger the first confirmation control, allowing the terminal to further display promotional information. For example, the preset position around the rendered image could be below or to the right of the rendered image; this application does not limit this.

[0105] In the above embodiments, by displaying a first confirmation control, the promotional information is pushed only after the user confirms. This avoids pushing invalid promotional information when the rendered image does not meet the user's needs and the user wants to change the description options, thus reducing resource waste.

[0106] In some exemplary embodiments, the promotional information display method of this application further includes: in response to a change operation of the selection of a description option under at least one description dimension, displaying a rendered image regenerated according to the currently selected description option under at least one description dimension; displaying a second confirmation control corresponding to the regenerated rendered image; and in response to a trigger operation of the second confirmation control, displaying promotional information corresponding to the regenerated rendered image.

[0107] The "selection change operation" refers to the operation of modifying the currently selected description option under the description dimension. For example, a selection change operation can specifically be the operation of changing the description option. For instance, under the shoe type dimension, if the currently selected description option is "Shoe Type 1," a selection change operation can select "Shoe Type 2," at which point "Shoe Type 1" is no longer selected. The selection change operation can also be the operation of adding or removing description options. For example, if two description options are currently selected under a certain description dimension, adding one more description option; or removing one of the selected description options; or deselecting a description option under a certain description dimension, so that description dimension no longer affects the generation of the rendered image, etc. This application does not specifically limit the selection change operation. The second confirmation control corresponding to the regenerated rendered image refers to an interactive component for confirming the rendered image. For example, it can be a button displayed corresponding to the regenerated rendered image and capable of responding to user triggering operations.

[0108] Specifically, if a user is dissatisfied with the displayed rendered image, they can re-trigger the first confirmation control corresponding to that rendered image and reselect the description options in each description dimension. This allows the user to trigger a regeneration of the rendered image through a selection change operation. The regenerated rendered image is generated according to the description option that was selected after the selection change operation, and the virtual objects contained in the regenerated rendered image match the description options that were selected after the selection change operation. The terminal can further display a second confirmation control at a preset position around the regenerated rendered image. When the user triggers the second confirmation control, promotional information corresponding to the regenerated rendered image is displayed. The promotional object described in this information matches the virtual objects in the regenerated rendered image.

[0109] Understandably, when a user is not satisfied with the regenerated rendered image, they can reselect it by making a change. Each time a user reselects, the terminal displays the rendered image generated according to the latest selected description option and the corresponding confirmation control. When the user confirms the rendered image through the confirmation control, it indicates that the rendered image matches the user's true intention. The server can then query promotional information based on the rendered image and push promotional information. The terminal can then display promotional information corresponding to the rendered image, and the promotional object described in the displayed promotional information matches the virtual object in the rendered image.

[0110] In the above embodiments, if the user is not satisfied with the currently rendered image, they can reselect the description options, which can trigger a re-rendering of the image, thus improving the flexibility of the promotional information display process.

[0111] In some exemplary embodiments, the rendered image is displayed in the image display area. The promotional information display method of this application further includes: displaying an image switching control in the image display area; in response to a trigger operation on the image switching control, displaying a historical rendered image in the image display area and displaying the description option corresponding to the historical rendered image as selected; displaying a third confirmation control corresponding to the historical rendered image, the third confirmation control being used to trigger the display of promotional information corresponding to the historical rendered image in response to a trigger operation on the third confirmation control.

[0112] In this context, an image switching control refers to an interactive component used to switch the images displayed in the image display area. In specific applications, an image switching control may include a first switching control for switching the previous rendered image and a second switching control for switching the next rendered image.

[0113] Specifically, in this embodiment, when multiple user selections trigger the generation and display of multiple rendered images, the terminal can also save all rendered images and display an image switching control in the image display area. When the user triggers the image switching control, historical rendered images can be displayed in the image display area, and the description option corresponding to the historical rendered image will be displayed as selected. The historical rendered image is the rendered image generated under multiple selections and whose generation time is before the current time. The terminal can also display a third confirmation control at a preset position around the historical rendered image. When the user triggers the third confirmation control, the terminal can display promotional information corresponding to the historical rendered image. The promotional object described in the promotional information matches the virtual object contained in the historical rendered image.

[0114] For example, if a user wants to search for sneakers, in the first selection, the user selects "Shoe Type 1 - Style 1 - Color Scheme 2". The terminal displays rendering image 1 generated according to "Shoe Type 1 - Style 1 - Color Scheme 2". Then, the user changes the selected "Style 1" to "Style 2" under the Style dimension and changes the selected "Color Scheme 2" to "Color Scheme 3" under the Color Scheme dimension. The terminal displays rendering image 2 generated according to "Shoe Type 1 - Style 2 - Color Scheme 3". At this point, rendering image 2 is the currently displayed rendering image, while rendering image 1 is a historical rendering image.

[0115] Understandably, when displaying historical rendered images, users can trigger the terminal to switch back to the previously displayed rendered image using the image switching control. For example, when rendering image 2 is displayed, the user can click the first switching control to switch to rendering image 1, and when rendering image 1 is displayed, the user can click the first switching control to switch to rendering image 2.

[0116] In the above embodiments, since the rendered images generated during multiple user selections can be saved, users can switch the displayed rendered images under different selection operations by triggering the image switching control, thereby facilitating better user decision-making and further improving the flexibility of the promotional information display process.

[0117] In some exemplary embodiments, the promotional information display method of this application further includes: displaying an image comparison entry; in response to a trigger operation on the image comparison entry, comparing and displaying the currently displayed rendered image with at least one historical rendered image; and in response to an image selection operation on the multiple rendered images displayed for comparison, displaying promotional information matching the rendered image selected by the image selection operation.

[0118] The image comparison entry point is the entry point that triggers image comparison. For example, the image comparison entry point can be an interactive component displayed at a preset position around the image display area. Comparison display refers to displaying multiple rendered images on the same page.

[0119] Specifically, the terminal can display an image comparison entry. When a user triggers the image comparison entry, the terminal can respond to the trigger and display an image comparison page. On this page, the currently displayed rendered image is compared with at least one historical rendered image. The user can compare and view the displayed rendered images and select the most satisfactory one. Then, in response to the image selection operation, the terminal can display promotional information that matches the rendered image selected by the image selection operation. The promotional object described in the promotional information matches the virtual object in the rendered image selected by the image selection operation.

[0120] For example, referring to Figure 5, the terminal can display an image comparison entry 502. When the user triggers the image comparison entry 502, the terminal can display an image comparison page 504. On the image comparison page 504, the currently displayed rendered image 2 is compared with the historical rendered image 1. The user can select the rendered image 1 and click the OK button. The terminal can then display promotional information that matches the rendered image 1.

[0121] In the above embodiments, since multiple rendered images can be compared and displayed, users can easily compare and view the rendered images generated by multiple selection operations, thereby selecting the most satisfactory rendered image more quickly and improving the interactive efficiency and flexibility in the process of displaying promotional information.

[0122] In some exemplary embodiments, the promotional information display method of this application further includes: when there are multiple description options under any description dimension, adding a preset mark to the popular description options among the multiple description options so that the popular description options are highlighted among the multiple description options.

[0123] Popular description options refer to those that receive attention, selection, or preference from most people.

[0124] Specifically, when multiple description options exist under any single description dimension, the terminal can identify the most popular description options among them. When displaying these multiple description options, a preset marker is added to the most popular description option to make it stand out. For example, "Recommended" or "Popular" can be added to the most popular description option. The terminal can determine the most popular description options based on data such as historical selection frequency and search popularity. For instance, description options ranking in historical selection frequency by a preset percentage (e.g., 20%) can be identified as most popular description options.

[0125] In the above embodiments, since preset markers can be added to popular description options, these popular description options will be highlighted among multiple description options. This allows users to quickly identify popular description options through the preset markers, enabling them to focus on the key points and improve interaction efficiency.

[0126] In some exemplary embodiments, the rendered image is a scene rendering image containing multiple virtual objects, displaying promotional information, including: displaying promotional information corresponding to each virtual object respectively; and displaying the identification information of the promotional information corresponding to each virtual object at the display position corresponding to each virtual object in the scene rendering image.

[0127] The scene rendering is a visualization that transforms a virtual 3D scene into a 2D image or animation, used to intuitively present the visual effects of a specific space, environment, or product. The scene rendering contains multiple virtual objects, each matching the description option selected by the user. The promotional information identification information is used to identify the promotional information, distinguishing each virtual object from the corresponding promotional information. This identification information can be, for example, Arabic numerals, alphabetical numbers, or the name of the virtual object corresponding to the promotional information.

[0128] Specifically, when the rendered image is a scene rendering image containing multiple virtual objects, the terminal can display promotional information corresponding to each virtual object. The promotional object described by each promotional information is matched with its corresponding virtual object. The terminal can further display the identification information of the promotional information corresponding to each virtual object at the display position of each virtual object in the scene rendering image.

[0129] In some exemplary embodiments, each identifier is used to identify a promotional message. Identifiers corresponding to the same virtual object are identified using different identifiers. When a user clicks on an identifier in the scene rendering, they are redirected to the details page of the promotional message identified by that identifier. On this details page, the user can browse a detailed description of the promoted object. The details page can also display transaction controls, which the user can trigger to initiate a transaction process for the promoted object. Identifying each promotional message with a different virtual object facilitates quick location of the specific promotional message for the user.

[0130] In some exemplary embodiments, each identification information is used to identify one or more promotional information items. Identification information items corresponding to the same virtual object can be identified using the same identification information. That is, one identification information is used to identify the set of all promotional information items corresponding to a virtual object. Using a unified identification information to identify the set of promotional information items corresponding to each virtual object can intuitively display the correspondence between identification information, promotional information items, and virtual objects.

[0131] For example, referring to Figure 6, when the user's search operation is for furniture, the terminal displays a scene rendering containing various types of furniture. For example, Figure 6 shows an interior scene rendering containing furniture such as chairs, sofas, and carpets, and displays promotional information for each piece of furniture. For example, in Figure 6, number 1 indicates three promotional items corresponding to the carpet, and number 2 indicates three promotional items corresponding to the sofa. Continuing to refer to Figure 6, in the scene rendering, number 2 is displayed corresponding to the sofa's display position, and number 1 is displayed corresponding to the carpet's display position.

[0132] In the above embodiments, the scene rendering image displays the identification information of the promotional information corresponding to each virtual object at the display position of each virtual object. By creating an immersive and seamless information acquisition path, the user's acceptance of the scene and promotional information can be improved. Through spatial information organization, the user's cognitive cost can be reduced and the efficiency of promotional information reach can be improved.

[0133] In some exemplary embodiments, the promotion information display method of this application further includes: in response to a trigger operation on the promotion information, displaying a details page corresponding to the promotion information; the details page displays a transaction control, which is used to respond to the trigger operation and trigger a transaction process for the promotion object described in the promotion information.

[0134] The details page corresponding to the promotional information refers to the in-depth display page that users enter through clicks, jumps, or other triggering operations. Its core function is to expand the "lightweight content" such as selling points and advantages extracted from the promotional information into a complete and detailed information set, helping users to fully understand the promotional object (such as products, services, activities, brands, etc.) and guide them to complete target behaviors (such as placing orders, registering, inquiring, etc.).

[0135] Specifically, users can trigger operations on the promotional information displayed on the terminal. The terminal can respond to the trigger operation by displaying the details page corresponding to the promotional information. On the details page, transaction controls can be displayed. When the user triggers the transaction controls, the transaction process for the promotional object described in the promotional information is initiated.

[0136] In practical applications, the terminal can also display more promotional information corresponding to the virtual object on the details page. On the details page of the virtual object, more related promotional information can be dynamically linked and displayed, such as similar promotional information, matching combinations, and derivative services, which can expand the user's interaction depth and choice space on the details page.

[0137] In the above embodiments, the details page can be accessed by triggering the promotional information, and transaction controls can be displayed on the details page. This allows users to conveniently and quickly perform interactive operations such as browsing and trading on the promoted objects, thus improving interaction efficiency.

[0138] In some exemplary embodiments, in response to a search operation, displaying description options for at least one description dimension of the search object of the search operation includes: in response to a search operation, displaying multiple dimension options of the search object of the search operation, where different dimension options correspond to different description dimensions; in response to a selection operation on multiple dimension options, displaying the selected dimension option; and displaying description options for the description dimension corresponding to the selected dimension option.

[0139] Specifically, the terminal can respond to a search operation by displaying multiple dimension options for the search object. Different dimension options correspond to different description dimensions, and the user can decide which description dimensions to use to describe the search object. The user can select one or more description dimensions from the multiple description dimensions displayed on the terminal, and the terminal can then display the selected dimension option. The terminal further displays the description options corresponding to the selected dimension option. Since the user can choose the description dimensions independently, the flexibility in the display of promotional information is improved.

[0140] In some exemplary embodiments, at least one description dimension includes a style dimension and a color scheme dimension. The rendered image is generated through an image rendering step, which includes: parsing the selected description option under the style dimension to obtain the style quantization parameter corresponding to the selected description option under the style dimension; mapping the selected description option under the color scheme dimension to a color space to obtain a color space parameter; and performing image rendering using an image rendering model based on the style quantization parameter and the color space parameter to obtain the rendered image.

[0141] Specifically, the selected description options under the style dimension are parsed to obtain style quantization parameters corresponding to the selected description options under the style dimension. Style quantization parameters can be, for example, line curvature, texture complexity, and light and shadow contrast. The selected description options under the color system dimension are mapped to a color space to obtain color space parameters. For example, the color system described by the selected description options under the color system dimension can be mapped to at least one color space, either RGB or HSV. Furthermore, a trained image rendering model is obtained. Based on the selected description options under each description dimension, a prompt word input style rendering model is constructed. The style quantization parameters and color space parameters are injected into the style rendering model as additional parameters, so that the style rendering model can render an image that can visually express the user's search intent.

[0142] In practical applications, the style rendering model can adopt the Stable Diffusion model, which is a text-to-image generation model that can generate a corresponding image based on the input text description.

[0143] Understandably, the above image rendering step can be performed by the terminal. That is, the terminal has an image rendering model installed. During the display of promotional information, the terminal generates a rendered image locally and then sends the rendered image, along with the semantic information expressed by the selected description options under each description dimension, to the server. The server can then query the promotional information based on the rendered image and the semantic information expressed by the selected description options under each description dimension. In other embodiments, the above image rendering step can be performed by the server. The terminal can send the semantic information expressed by the selected description options under each description dimension to the server. The server calls image rendering to generate a rendered image and then queries the promotional information based on the rendered image and the semantic information expressed by the selected description options under each description dimension.

[0144] In the above embodiments, by parsing the selected description options under the style dimension to obtain the style quantization parameters corresponding to the selected description options under the style dimension, mapping the selected description options under the color system dimension to the color space to obtain the color space parameters, and using the image rendering model to generate the rendered image, a more accurate rendered image can be obtained.

[0145] In some exemplary embodiments, promotional information is obtained through a promotional information acquisition step, which includes: performing multimodal feature analysis based on the rendered image and the selected description option to obtain feature vectors; calculating the matching degree between the feature vectors and each candidate promotional information in the promotional information database; and determining the promotional information from the promotional information database based on the matching degree corresponding to each candidate promotional information.

[0146] Multimodal feature analysis refers to the process of extracting and fusing features from different modalities of a rendered image and a selected description option. This includes extracting image features from the rendered image, extracting text features from the selected description option, aligning the extracted image features and text features across modalities, and then fusing the features to generate a feature vector for subsequent determination of promotional information.

[0147] Specifically, image features can be extracted from the rendered image first, and text features can be extracted from each selected description option. Then, cross-modal feature alignment is performed on the extracted image and text features. The aligned image and text features are then fused, and a feature vector is generated based on the fused features. The similarity between this feature vector and the feature vectors corresponding to each candidate promotional information in the promotional information database is calculated, thus obtaining the matching degree between the feature vector and each candidate promotional information. Based on the matching degree of each candidate promotional information, the promotional information is determined from the promotional information database. Specifically, the top M (M is a preset positive integer) candidate promotional information with the highest matching degree can be selected as the final determined promotional information.

[0148] In practical applications, pre-trained models (such as ResNet, VGG, and EfficientNet) can be used to extract high-level semantic features of images as extracted image features. For example, the 2048-dimensional visual feature vector output by the last fully connected layer of ResNet50 can be used as the extracted image features. Models such as Word2Vec, GloVe, or BERT can be used to extract text features of description options. The image feature vector and the text feature vector can be directly concatenated to form a high-dimensional joint feature. Weights are assigned according to modality confidence, such as by learning dynamic weights through a fully connected layer and then performing linear weighting to obtain fused features. Finally, an autoencoder is used to compress the fused high-dimensional features to obtain the feature vector.

[0149] In the above embodiments, cross-modal feature analysis can be performed, which effectively solves the problem of mismatch caused by feature ambiguity in single-modal retrieval and improves the consistency between promotional information and user intent.

[0150] In some exemplary embodiments, when the rendered image is a scene rendering image containing multiple virtual objects, the promotion information is determined from the promotion information database according to the matching degree corresponding to each candidate promotion information, including: determining the promotion information corresponding to each virtual object from the promotion information database according to the matching degree corresponding to each candidate promotion information; performing compatibility verification on the promotion information corresponding to each virtual object to obtain the compatibility verification result corresponding to each promotion information; and filtering out the promotion information whose corresponding compatibility verification result indicates that the verification failed from each promotion information.

[0151] Compatibility verification is the process of checking the suitability between different promotional objects involved in the promotional information. Compatibility verification ensures that different types of promotional objects meet physical size constraints or style consistency, such as the spacing threshold between a sofa and a coffee table meeting a threshold requirement. For physical size constraints, such as between a sofa and a coffee table, a pre-set spacing threshold range [L] is used. min ,L max ], where L min L represents the minimum spacing. max This represents the maximum spacing. Obtain the actual dimensions and placement information of the sofa and coffee table, calculate the actual spacing L between them, and if L... min ≤L≤L max If the physical size constraint is met, then the consistency is considered to be satisfied. For style consistency, the style feature vectors of the promoted objects can be extracted, and the similarity between the style feature vectors of different promoted objects can be calculated. If the similarity is greater than a preset similarity threshold S... threshold If so, then it is considered to satisfy style consistency.

[0152] Specifically, for each virtual object, the top N (N is a preset positive integer) promotional information items with the highest matching degree can be selected from the target promotional information set of the virtual object as the promotional information corresponding to the virtual object. Compatibility verification is performed on the promotional information corresponding to each virtual object to obtain the compatibility verification result of each promotional information item. When the compatibility verification result of a certain promotional information item indicates that the promotional object described by the promotional information item is incompatible with the promotional object described by any other promotional information item, it can be determined that the compatibility verification results of both promotional information items indicate that the verification has failed, and both promotional information items can be filtered.

[0153] For example, the remaining promotional information after filtering can be further verified for supply chain deliverability, that is, to verify whether the recommended information described in the promotional information is still in stock.

[0154] In the above embodiments, by performing compatibility verification on the promotional information obtained from the matching degree filtering, the reliability of the final displayed promotional information can be improved.

[0155] The above embodiments mainly introduce the specific implementation details of the promotion information display method on the front end. The following examples, taking the search object as the product and the description dimensions including style dimension and color system dimension, illustrate the implementation of the promotion information display method of this application in different application scenarios.

[0156] When searching for products, users may not be clear about what specific product they want to buy. Often, they need to adjust their keywords and search multiple times to see if the results meet their expectations. Referring to recommendation scenarios, they might look at real-time product image recommendations to determine their interest, indicating that users have visual preferences for products. These visual stimuli may further strengthen their liking for certain types of products. Based on this, this application proposes a method for searching products using visual information, providing users with a way to describe "the products they want" to find relevant products more quickly. Specifically, when a user performs a product search, information such as the user's search terms, search history, browsing history, and dwell time is analyzed to determine whether the user has the intention to "purchase" the product. For example, if a user's search terms contain keywords such as "purchase" or "place an order," or if the user spends a long time on the search results page and clicks on the product details page, then the user is considered to have the intention to "purchase" the product. When the intention to "purchase" the product is identified, the user is provided with related style and color scheme options based on their search terms to facilitate selection. Based on the user's selected combination parameters, AI generates 3D rendered product or scene images in real time. Based on the multimodal feature analysis of the generated rendered images and the text of the user's selected options, specific products in the product library are recalled and recommended to the user.

[0157] This application employs a positive feedback loop mechanism of "style semantic guidance → dynamic rendering generation → visual feedback optimization" to stimulate users' awareness of unconscious aesthetic preferences by generating guiding visual content in real time (such as preview images under different styles), allowing users to actively construct ideal product / scene visual images through selection. This application decouples style characteristics from brand, adopts a unified style description method, abstracts design elements into quantifiable parameters (such as line curvature and texture complexity), develops a supply chain-agnostic real-time inventory mapping engine, and supports dynamic replacement of alternative materials from multiple brands. In its implementation, this application also performs compatibility verification to ensure that recommended product combinations meet physical size constraints (such as the spacing threshold between sofas and coffee tables), style consistency, and supply chain deliverability (product inventory).

[0158] In some optional embodiments, this application provides an application scenario. Referring to Figure 7, in this application scenario, after a user enters a search term indicating their product intent, such as "sneakers," in the search box, a customized AI-generated and recommended shoe advertisement can be triggered. First, the user selects their preferred shoe style under the shoe style dimension, and a preview image can be displayed in the image display area below for the user to preview the selected shoe style. Based on the user's selected option, style options (such as "leopard print," "Y2K," etc.) suitable for that shoe style are provided to the user for selection again. Based on the user's selected option under the style dimension, color options suitable for that visual style are provided under the color scheme dimension. After the user selects a color scheme option, a preview image generated according to the selected shoe style option, style option, and color scheme option can be displayed in the image display area.

[0159] Referring to Figure 8, users can also perform selection change operations to reselect the description options under each description dimension, thereby changing the selected description options under each description dimension, and displaying a preview image generated according to the changed description options in the image display area. For example, in Figure 8, the selected description option under the style dimension is changed from "Y2K" to "leopard print", and the selected description option under the color scheme dimension is changed from the second to the third. Assuming that the user previously selected "shoe type 1" under the shoe type dimension, the preview image displayed in the image display area will be updated to a preview image generated according to shoe type 1, the changed style option, and the color scheme option.

[0160] Understandably, in this application scenario, the specific description dimensions selected and the options under each description dimension are not limited to this example. After the user selects the description option under the last dimension, they can click the "Start Customization" button to generate a rendered image of the product. Understandably, in specific applications, the color options under each description dimension also support dynamic expansion based on the user's selection hierarchy. For example, after the user selects the "Retro Style" option under the color dimension, sub-options such as "American Retro" and "Hong Kong Retro" may appear.

[0161] When a user clicks the "Start Customization" button in Figure 8, a rendered image is generated and displayed on the terminal. Referring to Figure 9, the preview image displayed in the image display area is updated to the rendered image of the product. After the rendered image is generated, multimodal feature analysis can be performed based on the image features of the rendered image and the text features of the description option selected by the user. Product data is then filtered and recalled in the product database for advertising recommendations, which are displayed side by side with the generated rendered image. Users can click on any of the displayed product advertisements to view product details.

[0162] In some alternative embodiments, this application also provides another application scenario. Referring to Figure 10, in this application scenario, a home decoration style image can be generated and suitable furniture products for the space can be recommended. Specifically, after a user searches for home decoration-related keywords such as "suitable furniture for Bohemian style decoration", the user can further select home preferences such as "loft", style tags such as "Bohemian style", and select a color scheme. After triggering the generation of a scene rendering image, a series of furniture products suitable for placement in the scene are recommended, and the product serial numbers are marked in the corresponding positions in the scene image.

[0163] The system architecture and technical details involved in the above application scenarios are explained below:

[0164] Referring to Figure 11, at the user layer, the target application of the terminal displays a user interface, which can display a search box, a multi-dimensional selection area, and an image display area. The terminal uses this user interface to realize user interaction, process user input and selection, and display AI-generated rendered images and recommended product advertisements.

[0165] Referring again to Figure 11, in the service layer, the server includes a user intent recognition module, a style management module, a dynamic color scheme recommendation module, an AI rendering engine, a cross-modal product matching module, and a real-time recommendation engine. The user intent recognition module analyzes user input to determine if there is an intent to purchase a product. The style management module maintains and provides a style semantic tag library. The dynamic color scheme recommendation module recommends dynamic color scheme options based on the selected style options. The AI ​​rendering engine generates 2D / 3D rendered images in real-time based on user-selected parameters. The cross-modal product matching module analyzes the visual features of the generated rendered images and the textual semantic features of the user-selected description options to perform cross-modal feature analysis. It then uses a product retrieval service to filter and retrieve relevant product data from the product database based on the matching results. The real-time recommendation engine dynamically adjusts product filtering strategies and ranking weights.

[0166] Referring again to Figure 11, in the data storage layer, there is a product database that stores product information, attributes, and inventory status. Optionally, the data storage layer also includes a user database to store user account information and interaction history, a style tag database to store style semantic tags and their relationships, and a rendering model library to store the basic models and materials used for AI generation.

[0167] In the AI ​​model layer, multiple AI models are set up, including a 3D rendering model, a visual-semantic joint mapping model, and a product compatibility verification model. The 3D rendering model is used to generate high-quality product or scene rendering images. The visual-semantic joint mapping model is used to establish the association between image features and text semantic features. The product compatibility verification model is used to ensure the rationality of recommended product combinations. The AI ​​model layer provides external AI services to the server through the set models. Referring to Figure 11, the server sends a rendering request through the AI ​​rendering engine to call the service and receives the rendering image returned by the external AI service.

[0168] Optionally, the system architecture can also include an external interface layer, which may include a supply chain management interface, a payment system interface, and a logistics system interface. Specifically, the supply chain management interface is used to connect with the merchant's inventory system to ensure the availability of recommended products, the payment system interface is used to process user purchase transactions, and the logistics system interface is used to manage product delivery.

[0169] Optionally, a security and monitoring layer can also be set in the system architecture to implement user authentication and authorization, encrypted data transmission, system performance monitoring, and logging.

[0170] Referring to Figure 12, which shows the interaction timing diagram of the technical solution of this application in some exemplary embodiments, the main interaction flow between the modules of this application is as follows:

[0171] 1. When a user enters search terms expressing their search intent through a product search operation on the terminal, the terminal sends a search request carrying the search terms to the user intent recognition module on the server.

[0172] 2. The user intent recognition module identifies the user's purchase intent and triggers the product customization process;

[0173] 3. The style management module and dynamic color scheme recommendation module provide users with description options under the dimensions of style and color scheme;

[0174] 4. The user makes a selection under each description dimension, and the terminal sends the user's selection results to the server's AI rendering engine. The user's selection results are the description options selected by the user under each description dimension.

[0175] 5. The AI ​​rendering engine generates a rendered image based on the user's selection and sends the rendered image to the terminal for display;

[0176] 6. The cross-modal matching module provides cross-modal matching services. It generates feature vectors by analyzing the rendered image and user selection results, and calls the product retrieval service to search for matching product data in the product database based on the feature vectors.

[0177] 7. The real-time recommendation engine sorts and filters the matched product data and sends it to the terminal;

[0178] 8. The terminal displays recommended product data, and users can view details or make purchases.

[0179] In summary, in the above application scenarios, the solution proposed in this application realizes the entire process from user input to personalized product recommendation, fully leveraging AI technology to improve user experience and recommendation accuracy. The beneficial effects of adopting the solution proposed in this application scenarios include at least the following:

[0180] 1) Visual Decision Assistance: Through the instant feedback of AI-generated images (including preview and rendered images), it helps users transform abstract concepts into concrete visual expressions, shortening the cognitive loop from intention formation to decision confirmation.

[0181] 2) Improve multi-scenario adaptability: Through a configurable style tag library and color rule engine, it supports rapid adaptation to the customization needs of different categories (such as clothing / home furnishings / digital products), reducing the technology migration cost of cross-industry applications.

[0182] 3) Mutual efficiency optimization: Through a hierarchical parameter selection mechanism, the complex decision-making process of users selecting products is decomposed into simplified steps that can be operated sequentially, reducing the cognitive load of users while ensuring the integrity of customized parameters.

[0183] 4) Improve the reliability of recommended products: Through a dual-path verification mechanism of visual feature analysis and semantic understanding, the problem of mismatch caused by feature ambiguity in single-modal retrieval is effectively solved, thereby improving the consistency between recommended products and user intent.

[0184] In some exemplary embodiments, this application also provides another method for displaying promotional information, which is executed by a computer device. The computer device may be the server 104 or the terminal 102 in FIG1, or it may be a system composed of the terminal 102 and the server 104. The following embodiments use the application of this method to the terminal 102 in FIG1 as an example for illustration.

[0185] It is understood that the promotional information display method provided in the following embodiments contains the same inventive concept as the promotional information display method provided in the foregoing embodiments. Therefore, the steps that are the same or similar to those in the foregoing embodiments in the following embodiments, the terms they contain and the specific implementation methods can be referred to the foregoing embodiments, and will not be repeated here.

[0186] It is also understood that, in the various embodiments mentioned above regarding the promotional information display method, embodiments that do not conflict with the promotional information display method provided in the embodiments below can be adaptively incorporated into the promotional information display method in the embodiments below, and all of them fall within the protection scope of this application. These embodiments will not be described in detail below.

[0187] Specifically, referring to Figure 13, the method for displaying promotional information includes the following steps:

[0188] Step 1302: Display promotional information and description options for at least one description dimension of the promotional object corresponding to the promotional information.

[0189] The promotional object corresponding to the promotional information is the promotional object described in the promotional information. There can be one or more promotional objects corresponding to the promotional information, with "multiple" meaning at least two. The description options for at least one description dimension of the promotional object refer to options used to describe the category to which the promotional object belongs from at least one description dimension. For example, if the promotional object is a brand of athletic shoes, then the description options for at least one description dimension of the promotional object could be options describing "athletic shoes" from at least one description dimension.

[0190] Specifically, the terminal can display promotional information, and corresponding to this promotional information, it can also display description options for at least one descriptive dimension of the promoted object. For example, the terminal can display promotional information, and corresponding to this information, display a first target control. This first target control is used to indicate related recommendations, for example, it could be a control indicating "Recommend Similar Products." When the user triggers the first target control, description options for at least one descriptive dimension of the promoted object corresponding to the promotional information are displayed. The first target control is an interactive component displayed along with the promotional information, and its function is to guide the user to perform related recommendation operations, commonly presented in the form of prompts such as "Recommend Similar Products."

[0191] Optionally, when there are many promotional targets corresponding to the promotional information, a second target control can be displayed. This second target control is used to indicate the filtering of promotional targets corresponding to the promotional information. For example, it could be a control indicating "Select your favorite product". When the user triggers the operation of the second target control, description options for at least one description dimension of the promotional targets corresponding to the promotional information are displayed. The second target control is an interactive control displayed in association with the promotional information when there are many promotional targets corresponding to the promotional information (determined by a preset quantity judgment condition). Its function is to trigger the operation of filtering promotional targets, and it usually guides the user to operate with prompts such as "Select your favorite product".

[0192] Step 1304: In response to the selection operation of description options under at least one description dimension, display the rendered image generated according to the selected description option under at least one description dimension.

[0193] Step 1306: Based on the rendered image, match at least one target promotion object and display it.

[0194] Here, the target promotion object refers to an object that matches the object described in the rendered image. In some embodiments, the target promotion object may be one of multiple promotion objects corresponding to the promotion information. In other embodiments, the target promotion object may be a promotion object other than the multiple promotion objects corresponding to the promotion information.

[0195] Specifically, the terminal can match and display at least one target promotional object based on the rendered image. For example, the terminal can display the target promotional object in the form of promotional information, such as an advertisement in any form, including images, text, or videos of the target promotional object.

[0196] For example, the terminal can send a rendered image to the server. The server then queries the promotion information database for a matching target promotion object based on the rendered image and returns the corresponding promotion information to the terminal for display. Optionally, the terminal can also send the user's selection results for each description dimension to the server, allowing the server to determine the target promotion object based on the rendered image and the selection results. The server first extracts features from the rendered image to obtain the image feature vector V. image Text feature extraction is performed on the selection results to obtain the text feature vector V. text V image and V text The fusion is performed to obtain the fused feature vector V. fusion Calculate V fusion The feature vector V of each candidate promotion object in the promotion information database candidate The similarity S between them can be expressed using the cosine similarity formula. The candidate promotion objects are sorted according to the similarity S, and the top N (N is a preset positive integer) candidate promotion objects are selected as the target promotion objects.

[0197] The following examples, using two application scenarios, illustrate the above-mentioned promotional information display method:

[0198] Scenario 1: The terminal displays an advertisement for a brand's products. The advertisement contains multiple similar products. To help users select their favorite product, the terminal can display a second target control. After the user clicks on this control, the terminal further displays description options under various description dimensions. After the user completes the selection, the terminal can display a rendered image. The user can then confirm the rendered image or change the selected option to regenerate the rendered image. The terminal can then display the target product that best matches the rendered image from among the multiple products in the advertisement.

[0199] Scenario 2: The terminal displays an advertisement for a certain brand's products. The user wants to be recommended more similar products from other brands. The user can then click on the second target control displayed on the terminal. The terminal further displays description options under various description dimensions. After the user completes the selection, the terminal can display a rendered image. The user can then confirm the rendered image or change the selected option to regenerate the rendered image. The terminal can then recommend and display more products similar to the product corresponding to the advertisement, thereby achieving related recommendations.

[0200] The aforementioned promotional information display method, by displaying promotional information and description options for at least one description dimension of the promotional object corresponding to the promotional information, can decompose the user's complex decision-making process into operable and simplified steps, reducing the user's cognitive load. Furthermore, in response to the selection operation of description options under at least one description dimension, a rendered image generated according to the selected description option under at least one description dimension is displayed. Thus, visual decision-making assistance can be formed through the real-time feedback of the rendered image, helping users to transform abstract needs into concrete visual expressions, shortening the cognitive loop from intention formation to decision confirmation. Furthermore, based on the rendered image, at least one target promotional object is matched and displayed, thereby enabling accurate recommendation of promotional objects that meet the user's needs based on the displayed promotional information.

[0201] In some exemplary embodiments, displaying promotional information and description options for at least one description dimension of the promotional object corresponding to the promotional information includes: displaying promotional information and description options for multiple description dimensions of the promotional object corresponding to the promotional information, arranged in a preset order; wherein, the description options displayed under each description dimension starting from the second description dimension are associated with the description option selected under the previous description dimension.

[0202] In some exemplary embodiments, in response to a selection operation of a description option under at least one description dimension, displaying a rendered image generated according to the selected description option under at least one description dimension includes: in response to a selection operation of a description option under at least one description dimension, displaying description options associated with the selected description option in a set of description options under the next description dimension; in response to a selection operation of at least one description option among the associated description options, if a next description dimension exists, performing the step of displaying description options associated with the selected description option in a set of description options under the next description dimension; and if a next description dimension does not exist, displaying a rendered image generated according to each of the selected description options.

[0203] In some exemplary embodiments, the promotional information display method further includes: in response to any description option changing from an initial state to a selected state, displaying a preview image generated according to the description option that has changed to the selected state, the preview image containing a preview object matching the description option that has changed to the selected state. It should be understood that although the steps in the flowcharts of the above embodiments are displayed sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the above embodiments may include multiple steps or multiple stages, which are not necessarily completed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least a portion of steps or stages in other steps.

[0204] Based on the same inventive concept, this application also provides a promotional information display device for implementing the promotional information display method described above. The solution provided by this device is similar to the solution described in the above method; therefore, the specific limitations in one or more promotional information display device embodiments provided below can be found in the limitations of the promotional information display method described above, and will not be repeated here.

[0205] In some embodiments, as shown in FIG14, a promotional information display device 1400 is provided, comprising:

[0206] Description options display module 1402 is used to display description options for at least one description dimension of the search object in response to a search operation;

[0207] The rendering image display module 1404 is configured to display a rendered image generated according to the selected description option under at least one description dimension in response to a selection operation of a description option under at least one description dimension, the rendered image containing a virtual object matching the selected description option.

[0208] The promotion information display module 1406 is used to display promotion information, and the promotion object described in the promotion information is matched with the virtual object.

[0209] The aforementioned promotional information display device, in response to a search operation, displays description options for at least one descriptive dimension of the search object, thereby breaking down the user's complex decision-making process into actionable, simplified steps, reducing the user's cognitive load. Furthermore, in response to the selection of a description option under at least one descriptive dimension, it displays a rendered image generated according to the selected description option. This rendered image contains virtual objects matching the selected description option. Thus, the instant feedback from the rendered image provides visual decision support, helping users transform abstract needs into concrete visual expressions, shortening the cognitive loop from intent formation to decision confirmation. The promotional information, which describes a promotional object that matches the virtual objects in the rendered image, effectively matches the user's search intent, thereby improving the promotional effect of the information and reducing the waste of promotional resources.

[0210] In some exemplary embodiments, the description option display module is further configured to: in response to a search operation, display description options of multiple description dimensions arranged in a preset order; wherein the description options displayed under each description dimension starting from the second description dimension are associated with the description option selected under the previous description dimension.

[0211] In some exemplary embodiments, the description option display module is further configured to: in response to a selection operation of a description option under at least one description dimension, display description options associated with the selected description option in the set of description options under the next description dimension; in response to a selection operation of at least one description option among the associated description options, if a next description dimension exists, perform the step of displaying description options associated with the selected description option in the set of description options under the next description dimension; if a next description dimension does not exist, display a rendered image generated according to each of the selected description options.

[0212] In some exemplary embodiments, the above apparatus further includes an image preview module, configured to display a preview image generated according to the description option that has been changed to the selected state in response to any description option changing from an initial state to a selected state, the preview image containing a preview object matching the description option that has been changed to the selected state.

[0213] In some exemplary embodiments, the rendering image display module is further configured to: display the selected description option under at least one description dimension in response to a selection operation of the description option under at least one description dimension; display a rendering trigger control when the selected description option under at least one description dimension satisfies the rendering trigger condition; and display a rendered image generated according to the selected description option under at least one description dimension in response to a trigger operation of the rendering trigger control.

[0214] In some exemplary embodiments, the promotional information display module is further configured to: display a first confirmation control corresponding to the rendered image in response to the display of the rendered image; and display promotional information in response to a triggering operation on the first confirmation control.

[0215] In some exemplary embodiments, the above apparatus further includes a promotional information update module, configured to: display a rendered image regenerated according to the currently selected description option under at least one description dimension in response to a change operation of the selection of description options under at least one description dimension; display a second confirmation control corresponding to the regenerated rendered image; and display promotional information corresponding to the regenerated rendered image in response to a trigger operation of the second confirmation control.

[0216] In some exemplary embodiments, the rendered image is displayed in the image display area, and the promotional information update module is further configured to: display an image switching control in the image display area; in response to a trigger operation on the image switching control, display a historical rendered image in the image display area and display the description option corresponding to the historical rendered image as selected; and display a third confirmation control corresponding to the historical rendered image, wherein the third confirmation control is used to trigger a trigger operation on the third confirmation control to trigger the display of promotional information corresponding to the historical rendered image.

[0217] In some exemplary embodiments, the promotion information update module is further configured to: display an image comparison entry; in response to a trigger operation on the image comparison entry, compare and display the currently displayed rendered image with at least one historical rendered image; and in response to an image selection operation on the multiple rendered images for comparison, display promotion information matching the rendered image selected by the image selection operation.

[0218] In some exemplary embodiments, the description option display module is further configured to: when there are multiple description options under any description dimension, add a preset mark to the popular description options among the multiple description options, so that the popular description options are highlighted among the multiple description options.

[0219] In some exemplary embodiments, the rendered image is a scene rendering image containing multiple virtual objects. The promotion information display module is further used to: display promotion information corresponding to each virtual object respectively; and display the identification information of the promotion information corresponding to each virtual object at the display position corresponding to each virtual object in the scene rendering image.

[0220] In some exemplary embodiments, the above-described apparatus further includes a details page display module, configured to: display a details page corresponding to the promotional information in response to a trigger operation on the promotional information; the details page displays transaction controls, which are used to trigger a transaction process for the promotional object described in the promotional information in response to the trigger operation.

[0221] In some exemplary embodiments, the description option display module is further configured to: in response to a search operation, display multiple dimension options of the search object of the search operation, wherein different dimension options correspond to different description dimensions; in response to a selection operation of multiple dimension options, display the selected dimension option; and display the description option of the description dimension corresponding to the selected dimension option.

[0222] In some exemplary embodiments, at least one description dimension includes a style dimension and a color scheme dimension. The rendered image is generated through an image rendering step. The above-mentioned apparatus further includes an image rendering module, which is used to parse the selected description option under the style dimension to obtain style quantization parameters corresponding to the selected description option under the style dimension; map the selected description option under the color scheme dimension to a color space to obtain color space parameters; and perform image rendering using an image rendering model based on the style quantization parameters and the color space parameters to obtain a rendered image.

[0223] In some exemplary embodiments, promotional information is obtained through a promotional information acquisition step. The above-mentioned apparatus further includes a promotional information acquisition module, which is used to: perform multimodal feature analysis based on the rendered image and the selected description option to obtain feature vectors; calculate the matching degree between the feature vectors and each candidate promotional information in the promotional information database; and determine the promotional information from the promotional information database based on the matching degree corresponding to each candidate promotional information.

[0224] In some exemplary embodiments, when the rendered image is a scene rendering image containing multiple virtual objects, the promotion information acquisition module is further configured to: determine the promotion information corresponding to each virtual object from the promotion information database according to the matching degree corresponding to each candidate promotion information; perform compatibility verification on the promotion information corresponding to each virtual object to obtain the compatibility verification result corresponding to each promotion information; and filter out the promotion information that fails the verification according to the corresponding compatibility verification result from each promotion information.

[0225] This application also provides another promotional information display device 1500, referring to Figure 15, including:

[0226] The promotion information display module 1502 is used to display promotion information and description options for at least one description dimension of the promotion object corresponding to the promotion information;

[0227] The rendering image display module 1504 is used to display a rendered image generated according to the selected description option under at least one description dimension in response to a selection operation of a description option under at least one description dimension.

[0228] The promotion object matching module 1506 is used to match at least one target promotion object based on the rendered image and display it.

[0229] In some exemplary embodiments, the promotion information display module is further configured to display promotion information and description options of multiple description dimensions of the promotion object corresponding to the promotion information, arranged in a preset order; wherein, the description options displayed under each description dimension starting from the second description dimension are associated with the description option selected under the previous description dimension.

[0230] In some exemplary embodiments, the above-described promotional information display device further includes: a preview module, configured to display a preview image generated according to the description option that has been changed to the selected state in response to any description option changing from an initial state to a selected state, the preview image containing a preview object matching the description option that has been changed to the selected state.

[0231] The aforementioned promotional information display device, by displaying promotional information and description options for at least one description dimension of the promotional object corresponding to the promotional information, can decompose the user's complex decision-making process into operable and simplified steps, reducing the user's cognitive load. Furthermore, in response to the selection operation of the description option under at least one description dimension, it displays a rendered image generated according to the selected description option under at least one description dimension. Thus, visual decision assistance can be formed through the real-time feedback of the rendered image, helping users to transform abstract needs into concrete visual expressions, shortening the cognitive loop from intention formation to decision confirmation. Furthermore, based on the rendered image, at least one target promotional object is matched and displayed, thereby enabling accurate recommendation of promotional objects that meet the user's needs based on the displayed promotional information.

[0232] Each module in the aforementioned promotional information display device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the operations corresponding to each module.

[0233] In some embodiments, a computer device is provided, which may be a server, and its internal structure diagram may be as shown in Figure 16. The computer device includes a processor, memory, input / output interfaces (I / O), and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is connected to the system bus via the I / O interfaces. The processor of the computer device provides computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database of the computer device stores promotional information. The I / O interfaces of the computer device are used for exchanging information between the processor and external devices. The communication interface of the computer device is used for communicating with external terminals via a network connection. When the computer program is executed by the processor, it implements a method for displaying promotional information.

[0234] In some embodiments, a computer device is provided, which may be a terminal, and its internal structure diagram may be as shown in Figure 17. The computer device includes a processor, memory, input / output interface, communication interface, display unit, and input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are connected to the system bus via the input / output interface. The processor of the computer device provides computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The input / output interface of the computer device is used for exchanging information between the processor and external devices. The communication interface of the computer device is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements a promotional information display method. The display unit of the computer device is used to form a visually visible image. It can be a display screen, a projection device, or a virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0235] Those skilled in the art will understand that the structures shown in Figures 16 and 17 are merely block diagrams of some structures related to the present application and do not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than shown in the figures, or combine certain components, or have different component arrangements.

[0236] In some embodiments, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the above-described promotional information display method.

[0237] In some embodiments, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described promotional information display method.

[0238] In some embodiments, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps of the above-described promotional information display method.

[0239] In summary, this application provides a method, apparatus, computer device, computer-readable storage medium, and computer program product for displaying promotional information. By responding to search operations, it displays description options for at least one descriptive dimension of the search object, presenting information in a structured manner. This reduces user cognitive complexity, decreases the time and effort required for users to understand search conditions, and improves the efficiency of human-computer interaction. Furthermore, by responding to description option selection operations and displaying rendered images, it utilizes the real-time feedback of artificial intelligence technology to transform abstract user needs into visual virtual objects. This allows the system to more accurately capture user intent, reduces erroneous matching caused by information misunderstanding, improves the accuracy and targeting of subsequent promotional information recommendations, avoids ineffective distribution of promotional resources, and increases resource utilization.

[0240] Furthermore, in response to search operations, multiple descriptive options with related dimensions are displayed in a preset order. This hierarchical information display structure conforms to human cognitive logic, reduces information overload when users have to choose from a large number of options, improves the accuracy and efficiency of selection operations, reduces the probability of users making incorrect selections, and enables the system to process user choices more efficiently, thereby improving the system's processing efficiency.

[0241] When responding to a description option selection action, the description options associated with the selected description option are displayed progressively until the rendered image is displayed. This dynamic filtering and guidance method reduces the user's search range during the selection process, improves selection efficiency, and ensures that the rendered image can more accurately reflect the user's final needs, thereby improving the accuracy and quality of the generated rendered image.

[0242] When the description option status changes, a preview image is displayed, providing users with real-time visual feedback. This allows users to adjust their selections promptly, reducing repetitive operations due to incorrect choices and improving the accuracy and efficiency of user operations. At the same time, it helps the system more accurately capture users' potential aesthetic preferences, providing richer evidence for subsequent promotional information matching and improving matching accuracy.

[0243] The response description option selection operation displays the selected description option, and the rendering trigger control is displayed when the rendering trigger condition is met. This interactive design gives users the right to control the generation of rendered images, avoids unnecessary image rendering operations, reduces the consumption of system resources, and improves the flexibility and resource utilization of the rendered image generation process.

[0244] When displaying promotional information, the user confirms the rendered image through a first confirmation control before the promotional information is displayed. This avoids invalid promotional information queries and pushes when the rendered image does not meet the user's needs, reduces the system's computing resource consumption and network bandwidth usage, improves resource utilization, avoids interference with the user, and enhances the user experience.

[0245] The system responds to changes in the description options by regenerating the rendered image and displaying a confirmation control. This mechanism allows users to adjust their selections according to the actual situation, improving the system's flexibility and adaptability. It ensures that promotional information accurately matches the user's final needs, thereby enhancing the accuracy and effectiveness of the promotional information.

[0246] When a rendered image is displayed in the image display area, an image switching control is shown, allowing users to easily view historical rendered images. This helps users compare the effects of different selections, make more accurate decisions, reduce the time users spend re-selecting and re-rendering, improve decision-making efficiency, and also enhance the system's flexibility and user experience.

[0247] The system displays an image comparison entry, compares and renders images, and shows matching promotional information based on the user's selection. This provides users with a more intuitive means of comparison, reduces the decision-making time when users choose from multiple options, improves interaction efficiency and decision accuracy, and also helps the system to more accurately meet user needs.

[0248] When there are multiple description options in the description dimension, preset tags are added to popular description options. By leveraging users' herd mentality and search habits, the time users spend searching for key points among many options is reduced, improving interaction efficiency. At the same time, it also helps the system guide users to focus on popular options, thus optimizing the system's recommendation algorithm performance.

[0249] When the rendered image is a scene rendering image, promotional information and identification information corresponding to the virtual objects are displayed respectively. Through spatial information organization, the user's cognitive cost is reduced and the efficiency of users obtaining information is improved. At the same time, the identification information establishes a clear association between promotional information and virtual objects, thereby improving the efficiency and accuracy of promotional information delivery.

[0250] The system responds to promotional information to trigger actions that display details pages and transaction controls, providing users with a one-stop browsing and transaction experience. This reduces the time and operational costs for users to switch between different pages, improves interaction efficiency, and also facilitates the system's processing of user transaction requests, thereby improving the system's processing efficiency and conversion rate.

[0251] The system displays multiple dimension options in response to search operations and allows users to choose them independently, enhancing the system's flexibility and customizability. This enables users to select appropriate description dimensions based on their own needs, reducing the resource consumption of displaying unnecessary description dimensions and improving resource utilization. It also better meets users' personalized needs.

[0252] When at least one description dimension includes style and color scheme dimensions, by quantifying style and color scheme information and using an image rendering model to generate rendered images, the accuracy and consistency of rendered image generation are improved, image deviations caused by differences in subjective understanding are reduced, the system can more accurately present user needs, and the user experience is improved.

[0253] Promotional information is determined through multimodal feature analysis, which combines the visual features of the rendered image and the semantic features of the description options. This effectively solves the mismatch problem caused by feature ambiguity in single-modal retrieval, improves the consistency between promotional information and user intent, and enhances the robustness and accuracy of the recommendation system.

[0254] When the rendered image is a scene rendering image, compatibility verification is performed on the promotional information and incompatible information is filtered out. This ensures the coordination between recommended promotional objects in terms of physical size and style, improves the reliability of recommendation results and user satisfaction, and reduces user complaints and system processing costs caused by incompatible promotional objects.

[0255] Another method for displaying promotional information involves showing description options for the promotional information and corresponding target audience, guiding users to express their needs in a hierarchical manner. This reduces the cognitive load on users and improves the system's efficiency in capturing user needs. Responding to the selection of description options displays a rendered image, assisting users in decision-making through visual feedback. This transforms abstract user needs into concrete visual expressions, improving the accuracy and efficiency of matching target audiences. Matching and displaying target audiences based on rendered images achieves precise recommendations based on user needs, improving the relevance and targeting of promotional information, reducing the display of invalid promotional information, and increasing resource utilization and user conversion rates.

[0256] Furthermore, the system displays description options arranged in a preset order and related to each other, avoiding users being faced with a massive number of filtering conditions, reducing cognitive complexity, improving the interactive efficiency of selection operations, enabling the system to process user input more efficiently, and improving the system's response speed.

[0257] When the description option status changes, a preview image is displayed to provide users with real-time feedback, helping them to adjust their choices in a timely manner, reducing the probability of incorrect selections, improving interaction efficiency, and also helping the system to more accurately understand the user's potential needs and optimize the effect of the recommendation algorithm.

[0258] In terms of system architecture, a layered design is adopted, including a user layer, a service layer, a data storage layer, and an AI model layer. The user layer achieves efficient human-computer interaction through a user-friendly interface, improving the convenience and accuracy of user operations. Multiple modules in the service layer collaborate, such as the user intent recognition module accurately judging user needs, reducing errors caused by misjudgments; the style management module and dynamic color recommendation module provide rich and relevant descriptive options, enhancing the system's guidance and the effectiveness of user choices; the AI ​​rendering engine utilizes real-time generation technology, improving the speed and quality of rendered images and reducing user waiting time; the cross-modal product matching module and real-time recommendation engine combine multimodal feature analysis and dynamic filtering strategies to improve the accuracy and timeliness of product recommendations. The data storage layer provides stable and reliable data support for the system, ensuring efficient data storage and fast retrieval, reducing data access latency. Multiple models in the AI ​​model layer work collaboratively: the 3D rendering model generates high-quality rendered images, the visual-semantic joint mapping model establishes the association between image and text features, and the product compatibility verification model ensures the rationality of recommended products, improving the intelligence and reliability of the entire system. The setup of the external interface layer and security monitoring layer further enhances the system's scalability, security, and stability, reduces interference and impact from external factors, and ensures the normal operation of the system and data security.

[0259] From the perspective of interaction sequence, the modules interact in an orderly manner according to a predetermined process, reducing communication delays and errors between modules and improving the overall response speed and processing efficiency of the system. Through standardized interaction processes, the system can quickly respond to user operations, promptly generate and recommend promotional information that meets user needs, thereby enhancing user experience and satisfaction.

[0260] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data shall comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0261] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, data processing logic devices, etc., and are not limited to these.

[0262] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0263] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for displaying promotional information, executed by a computer device, the method comprising: In response to a search operation, display description options for at least one description dimension of the search object for the search operation; In response to a selection operation of a description option under the at least one description dimension, a rendered image generated according to the selected description option under the at least one description dimension is displayed, the rendered image containing a virtual object matching the selected description option; and The promotional information is displayed, and the promotional object described in the promotional information matches the virtual object.

2. The method according to claim 1, wherein displaying description options for at least one description dimension of the search object in response to a search operation includes: In response to a search operation, display description options for multiple description dimensions arranged in a preset order; In this context, the description options displayed under each description dimension, starting from the second description dimension, are associated with the description option selected under the previous description dimension.

3. The method according to claim 2, wherein displaying a rendered image generated according to the selected description option under the at least one description dimension in response to a selection operation of a description option under the at least one description dimension includes: In response to a selection operation of a description option under the at least one description dimension, display the description options associated with the selected description option in the set of description options under the next description dimension; In response to a selection operation of at least one of the associated description options, if a next description dimension exists, the step of displaying the description options associated with the selected description option in the set of description options under the next description dimension is performed; If no next description dimension exists, display the rendered image generated according to the selected description options.

4. The method according to any one of claims 1 to 3, further comprising: In response to any description option changing from an initial state to a selected state, a preview image generated according to the description option that has changed to the selected state is displayed, the preview image containing a preview object that matches the description option that has changed to the selected state.

5. The method according to any one of claims 1 to 4, wherein displaying a rendered image generated according to the selected description option under the at least one description dimension in response to a selection operation of a description option under the at least one description dimension comprises: In response to a selection operation of a description option under the at least one description dimension, the selected description option under the at least one description dimension is displayed; When the selected description option under at least one description dimension meets the rendering trigger condition, the rendering trigger control is displayed; In response to a trigger operation on the rendering trigger control, a rendered image generated according to the selected description option under the at least one description dimension is displayed.

6. The method according to any one of claims 1 to 5, wherein displaying promotional information comprises: In response to the display of the rendered image, a first confirmation control corresponding to the rendered image is displayed; In response to the triggering operation of the first confirmation control, promotional information is displayed.

7. The method according to claim 6, further comprising: In response to a change operation on the selection of description options under the at least one description dimension, a rendered image regenerated according to the currently selected description option under the at least one description dimension is displayed; Display a second confirmation control corresponding to the regenerated rendered image; In response to the triggering operation of the second confirmation control, promotional information corresponding to the regenerated rendered image is displayed.

8. The method according to claim 7, wherein the rendered image is displayed in the image display area, the method further comprising: An image switching control is displayed in the image display area; In response to a trigger operation on the image switching control, a historical rendered image is displayed in the image display area, and the description option corresponding to the historical rendered image is displayed as selected. Display a third confirmation control corresponding to the historical rendered image. The third confirmation control is used to respond to a trigger operation to trigger the display of promotional information corresponding to the historical rendered image.

9. The method according to claim 7 or 8, further comprising: Display image comparison entry; In response to a trigger operation on the image comparison entry, the currently displayed rendered image is compared and displayed with at least one historical rendered image; In response to an image selection operation on multiple rendered images displayed for comparison, promotional information matching the rendered image selected by the image selection operation is displayed.

10. The method according to any one of claims 1 to 9, wherein the method further comprises: When there are multiple description options under any description dimension, a preset mark is added to the popular description options among the multiple description options so that the popular description options are highlighted among the multiple description options.

11. The method according to any one of claims 1 to 10, wherein the rendered image is a scene rendering image containing multiple virtual objects, and the display of promotional information includes: Display the promotional information corresponding to each virtual object; In the scene rendering image, corresponding to the display position of each virtual object, the identification information of the promotional information of each virtual object is displayed.

12. The method according to any one of claims 1 to 11, wherein the method further comprises: In response to a triggering operation on the promotional information, the details page corresponding to the promotional information is displayed; The details page displays transaction controls, which are used to respond to trigger operations and initiate a transaction process for the promoted object described in the promotional information.

13. The method according to any one of claims 1 to 12, wherein displaying description options for at least one description dimension of the search object of the search operation in response to a search operation comprises: In response to a search operation, multiple dimension options for the search object are displayed, with different dimension options corresponding to different descriptive dimensions; In response to the selection operation of the multiple dimension options, the selected dimension option is displayed; Displays the description options for the selected dimension.

14. The method according to any one of claims 1 to 13, wherein the at least one description dimension includes a style dimension and a color scheme dimension, and the rendered image is generated through an image rendering step, the image rendering step comprising: The selected description options under the style dimension are parsed to obtain the style quantization parameters corresponding to the selected description options under the style dimension; Map the selected description options under the color system dimension to the color space to obtain the color space parameters; Based on the style quantization parameters and the color space parameters, an image rendering model is used to render the image and obtain the rendered image.

15. The method according to any one of claims 1 to 14, wherein the promotional information is obtained through a promotional information acquisition step, the promotional information acquisition step comprising: Multimodal feature analysis is performed based on the rendered image and the selected description option to obtain feature vectors; Calculate the matching degree between the feature vector and each candidate promotion information in the promotion information database; Promotional information is determined from the promotional information database based on the matching degree of each candidate promotional information.

16. The method according to claim 15, wherein when the rendered image is a scene rendering image containing multiple virtual objects, determining the promotion information from the promotion information database based on the matching degree corresponding to each candidate promotion information includes: Based on the matching degree of each candidate promotion information, the promotion information corresponding to each virtual object is determined from the promotion information database. Perform compatibility verification on the promotional information corresponding to each virtual object, and obtain the compatibility verification results for each piece of promotional information. From each promotional message, filter out those that failed the compatibility verification.

17. A method for displaying promotional information, executed by a computer device, the method comprising: Display promotional information, and description options for at least one description dimension of the promotional object corresponding to the promotional information; In response to a selection operation of a description option under the at least one description dimension, a rendered image generated according to the selected description option under the at least one description dimension is displayed; and Based on the rendered image, match at least one target promotion object and display it.

18. The method according to claim 17, wherein displaying promotional information and description options for at least one description dimension of the promotional object corresponding to the promotional information includes: Display promotional information, as well as description options for multiple description dimensions of the promotional object corresponding to the promotional information, arranged in a preset order; In this context, the description options displayed under each description dimension, starting from the second description dimension, are associated with the description option selected under the previous description dimension.

19. The method according to claim 17 or 18, further comprising: In response to any description option changing from an initial state to a selected state, a preview image generated according to the description option that has changed to the selected state is displayed, the preview image containing a preview object that matches the description option that has changed to the selected state.

20. A promotional information display device, the device comprising: A description option display module is used to display description options for at least one description dimension of the search object in response to a search operation; A rendering image display module is configured to, in response to a selection operation of a description option under the at least one description dimension, display a rendered image generated according to the selected description option under the at least one description dimension, the rendered image containing a virtual object matching the selected description option; The promotion information display module is used to display promotion information, wherein the promotion object described in the promotion information matches the virtual object.

21. A promotional information display device, the device comprising: The promotion information display module is used to display promotion information and description options for at least one description dimension of the promotion object corresponding to the promotion information; A rendering image display module is configured to display a rendered image generated according to the selected description option under the at least one description dimension in response to a selection operation of the description option under the at least one description dimension. The promotion target matching module is used to match at least one target promotion target based on the rendered image and display it.

22. A computer device comprising a memory and a processor, the memory storing a computer program, the processor executing the computer program to implement the steps of the method according to any one of claims 1 to 19.

23. A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method according to any one of claims 1 to 19.

24. A computer program product comprising a computer program that, when executed by a processor, implements the steps of the method according to any one of claims 1 to 19.