Wallpaper display method, terminal device, readable medium and program product

Abstract

The present invention relates to the technical field of terminals, and specifically relates to a wallpaper display method, a terminal device, a readable medium and a program product. In the method, by means of adding wallpaper elements in multiple dimensions, such as attire, makeup and wearable devices, to foreground elements included in a user-specified wallpaper material image, adjusting poses or replacing background elements, and performing final processing, personalized target wallpaper is obtained and displayed. The pose and the background elements can also be referred to as wallpaper elements in different dimensions. In the generated target wallpaper, the foreground elements can be matched with attire, makeup, wearable devices, etc., that a user likes, the body poses of the background elements can also be adjusted, and the background elements can be blended with background elements that the user likes. In this way, the user requirements of personalized wallpaper customization and usage can be satisfied.

Description

Wallpaper display method, terminal device, readable medium and program product

[0001] The present application claims priority to the Chinese patent application No. 202411002270.6, filed on July 24, 2024, and entitled "Wallpaper display method, terminal device, readable medium and program product", the content of which is incorporated herein by reference in its entirety. TECHNICAL FIELD

[0002] The present application relates to the technical field of terminal, in particular to a wallpaper display method, a terminal device, a readable medium and a program product. BACKGROUND

[0003] With the development of terminal technology, in order to meet the various personalized needs of users, mobile phones and other terminal devices also gradually support more and more personalized customization, such as supporting users to set personalized avatars or personalized wallpapers, etc.

[0004] However, the current wallpaper display scheme can only support users to select images downloaded or taken by themselves, which are only set as wallpapers after simple cropping, blurring and other processing, and cannot meet the more personalized customization needs of users. SUMMARY

[0005] The present application provides a wallpaper display method, a terminal device, a readable medium and a program product, which can meet the personalized customization and use needs of users for wallpapers, recommend the desired goods to users through personalized wallpapers, recommend the generated personalized wallpapers to applications for use as picture materials, provide more convenient services for users, and is beneficial to improve the use experience of users.

[0006] In a first aspect, the present application provides a wallpaper display method applied to a terminal device, the method comprising: detecting a first input operation of a user, wherein the first input operation is used to specify a first image as a generation source of a wallpaper, the first image comprising a first person with a first facial feature, and the first person has a first display state in the first image, the first display state comprising at least one of clothing, makeup, wearable devices, background, and style; and displaying a second image as a wallpaper, wherein the second image is generated based on the first input operation and the first image, the second image comprising the first person with the first facial feature, and the first person has a second display state in the second image, the second display state comprising at least one of clothing, makeup, wearable devices, background, and style, and the first display state and the second display state are different.

[0007] For example, the first image can be a wallpaper material image specified by a user, i.e., the image can be an object for setting a wallpaper by the user. Correspondingly, the first input operation can include an operation for setting a wallpaper based on an image displayed by a terminal device such as a mobile phone, i.e., the first input operation can specify the first image as a source for generating a wallpaper, i.e., a wallpaper material image. The first image as a source for generating a wallpaper can provide a first face feature corresponding to a first person, and face control information corresponding to the first face feature can control the finally generated wallpaper, i.e., the second image also has the same person. Therefore, the generated second image, e.g., a personalized wallpaper, can include the first person with the first face feature.

[0008] Based on this, in the case of controlling the first person to be unchanged based on the first face feature, the first person in the first image, i.e., the wallpaper material image, can be added with wallpaper elements of different dimensions, e.g., one or more of clothing, makeup, wearable devices, backgrounds, styles, to change the different display states of the first person, and finally process a personalized wallpaper required by the user, i.e., the second image. In this way, the above scheme can meet the personalized customization and use requirements of the user for the wallpaper.

[0009] In a possible implementation of the first aspect, the first input operation is used to specify the second display state.

[0010] For example, the user can specify the first image as a source for generating a wallpaper through a relevant wallpaper setting interface, and can specify parameters, information identifiers, or images capable of providing relevant elements of the first input operation to set relevant controls, to specify different dimensions of the first person in the first image, e.g., clothing, makeup, wearable devices, backgrounds, styles, etc.

[0011] In a possible implementation of the first aspect, the second display state of the first person in the second image includes a first part state and a second part state, and the second image is generated based on the first input operation and the first image, including: the second image is obtained by processing the first image according to the first part state and the second part state, wherein the first part state is determined based on the first input operation, and the second part state is determined based on a default setting.

[0012] In some embodiments, the manner in which the first character with the first facial feature changes the reality state can be matched or adjusted based in part on default settings of the system and in part on parameters, information identifiers, and the like of the relevant element image or the relevant control specified by the user. In other embodiments, the manner in which the first character with the first facial feature changes the reality state can be matched or adjusted based entirely on default settings of the system, for example, after the user specifies the first image as the generation source, the terminal device such as a mobile phone can match the display state of the first character based on user behavior data collected during the running of some associated applications or based on random or sequential matching and corresponding adjustment of the display state of the first character according to the system default settings, without limitation.

[0013] In a possible implementation of the first aspect, the first display state and the second display state are different, including that the first character in the first image has a first style and the first character in the second image has a second style.

[0014] In a possible implementation of the first aspect, the first input operation is used to specify the second style in the second display state, and the second image is generated based on the first input operation and the first image, including that the second image is obtained by processing the first image according to prompt information corresponding to the second style, wherein the prompt information corresponding to the second style is generated based on a style parameter corresponding to the second style.

[0015] For example, for the style of the first character in the first image, including the style corresponding to the clothing or outfit, the style corresponding to the scenery, and the like, the corresponding information identifier of the wallpaper style specified by the first input operation of the user can be used to obtain the corresponding style parameter to generate the corresponding prompt information to input the diffusion model for model inference, and finally generate the personalized wallpaper conforming to the style required by the user.

[0016] Similarly, for the clothing of the first character in the first image, the corresponding clothing control information can also be obtained based on the clothing element image or the outfit template specified by the first input operation of the user, and also used as input information of the relevant diffusion model. For the makeup of the first character in the first image, the corresponding makeup control information can also be obtained based on the makeup element image or the makeup template specified by the first input operation of the user, and also used as input information of the relevant diffusion model. For the wearable device worn by the first character in the first image, the corresponding device control information can also be obtained based on the device element image or the device template specified by the first input operation of the user, and also used as input information of the relevant diffusion model. For the background of the first character in the first image, the corresponding background control information can also be obtained based on the background element image or the background template specified by the first input operation of the user, and also used as input information of the relevant diffusion model, without limitation.

[0017] In a possible implementation of the first aspect, the first display state further includes a posture of the first character, and the first display state is different from the second display state, including: the first posture that the first character has in the first image is different from the second posture that the first character has in the second image.

[0018] In a possible implementation of the first aspect, the first input operation is used to specify the second posture in the second display state, and the second image is generated based on the first input operation and the first image, and further including: the second image is obtained by processing the first character in the first image according to a posture feature vector corresponding to the second posture, wherein the posture feature vector corresponding to the second posture is obtained based on the first posture template; or the posture feature vector corresponding to the second posture is obtained by extracting feature information based on a third image specified by the user.

[0019] For example, for the posture of the first character in the first image, the corresponding posture control information can also be obtained based on the posture element image or the posture template specified by the first input operation of the user, and used as input information of the related diffusion model.

[0020] In a possible implementation of the first aspect, the method further includes: sending image data corresponding to the first image to a server; receiving the second image returned by the server, and displaying the second image as a wallpaper.

[0021] For example, a terminal device such as a mobile phone can upload the first image specified by the user, that is, a wallpaper material image, to a server for processing into a personalized wallpaper, and then obtain the generated personalized wallpaper from the server for display. In this way, when the terminal device such as a mobile phone has weak processing capability, the customization and use demand of the user for the personalized wallpaper can also be met.

[0022] In a possible implementation of the first aspect, the method includes: displaying a third image as a wallpaper, wherein the third image is obtained by adding an identifier of a product associated with the second display state to the second image.

[0023] That is, the second image obtained by the above processing, for example, the personalized wallpaper, the wallpaper elements related to the state of the first character in the second display state, such as clothing, makeup, wearable devices, and background, can also be associated with the relevant product link to recommend to the user for purchase. In this way, the personalized wallpaper can recommend the desired product to the user, provide more convenient services for the user, and improve the user experience.

[0024] In some other embodiments, the second image obtained by the above processing, for example, the personalized wallpaper, can also be recommended to the application as a picture material for use, which is also conducive to improving the user experience.

[0025] In a possible implementation of the first aspect, the displaying the third image as the wallpaper includes: in response to a second input operation of the user, switching the displayed wallpaper from the second image to the third image.

[0026] In a possible implementation of the first aspect, the second input operation includes an operation of the user clicking or selecting a region around at least one wallpaper element in the second image.

[0027] For example, the user can click a part of the first person in the second image, for example, the personalized wallpaper, such as the clothing, makeup, wearable device, background, etc., or can select a region in the second image, for example, the personalized wallpaper, such as the clothing, makeup, wearable device, background, etc., by a region screenshot or screen capture operation. In a possible implementation of the first aspect, the method further includes: in response to an operation of the user identifying the commodity, displaying a recommendation interface of the commodity.

[0028] The operation of the user clicking or selecting a region around, that is, the second input operation, can indicate that the user is interested in the related commodity, such as the clothing, makeup, wearable device, background, etc., in the personalized wallpaper. Therefore, based on the operation, the identification of the associated commodity is displayed on the personalized wallpaper, which can facilitate the user to view or purchase the related commodity in the related shopping application, provides more convenient services for the user, and is conducive to improving the user experience.

[0029] In some other embodiments, the user operation indicating that the user is interested in the wallpaper element in the personalized wallpaper can also include an operation of the user clicking a collection, screen capture or screenshot, and downloading or saving the target wallpaper image on the related interface, for example, the lock screen interface, and an operation of the user staying on the related interface for more than a preset time threshold, etc., which is not limited herein.

[0030] In a possible implementation of the first aspect, the recommendation interface of the commodity is an interface of a first application installed on the terminal device. For example, the first application can be a shopping application. Based on the wallpaper display method provided in the present application, the recommendation interface of the related commodity provided by the shopping application can be displayed based on the second input operation of the user while displaying the personalized wallpaper. Based on this, the identification of the commodity can correspond to an associated commodity link, and the commodity link can be provided by the shopping application.

[0031] In a possible implementation of the first aspect, the commodity includes one or more of the following: corresponding to a clothing-related wallpaper element, the commodity includes any of a hat, a piece of clothing, a pair of shoes, a necklace, a bracelet, and ear jewelry; corresponding to a makeup-related wallpaper element, the commodity includes any of an eyebrow pencil, an eye shadow, a lipstick, a toner, and a cream; corresponding to a wearable device-related wallpaper element, the commodity includes any of a watch, a headset, and glasses; corresponding to a background-related wallpaper element, the commodity includes a travel product related to a scenic spot corresponding to the background.

[0032] In a possible implementation of the first aspect, the first input operation is an operation of setting a wallpaper by the user, or the first input operation is an interface operation of the user on a second application installed on the terminal device.

[0033] For example, the second application can be, for example, a shopping application or a media application browsed by the user. For example, when the user browses a desired picture in the shopping application or the media application, the user can perform an operation of setting the desired picture as a wallpaper, that is, the first input operation, thereby triggering the terminal device such as a mobile phone to perform the wallpaper display method provided in this application to display the corresponding generated personalized wallpaper.

[0034] In a possible implementation of the first aspect, the method further includes recommending the generated second image to a third application.

[0035] For example, the personalized wallpaper generated based on the wallpaper display method provided in this application can be recommended to a shopping application as a buyer show or the like for use by a buyer or a seller, so that the commercial value of the personalized wallpaper can be improved.

[0036] In a second aspect, the application provides a wallpaper display method, including: a terminal device detecting a first input operation of a user, wherein the first input operation is used to specify a first image as a generation source of a wallpaper, the first image including a first character with a first facial feature, and the first character has a first display state in the first image, and the first display state includes at least one of clothing, makeup, a wearable device, a background, and a style; the terminal device sending operation data corresponding to the first input operation and the first image to a server; the server generating a second image based on the operation data corresponding to the first input operation and the first image, wherein the second image includes the first character with the first facial feature, the first character has a second display state in the second image, the second display state includes at least one of clothing, makeup, a wearable device, a background, and a style, and the first display state and the second display state are different; and the terminal device receiving the second image returned by the server and displaying the second image as a wallpaper.

[0037] In a third aspect, the present application provides a terminal device, comprising: one or more processors; one or more memories; the one or more memories storing one or more programs, which, when executed by the one or more processors, cause the terminal device to perform the wallpaper display method provided in the first aspect and the various possible implementations of the first aspect.

[0038] In a fourth aspect, the present application provides a computer readable medium, which stores instructions, which, when executed on a computer, cause the computer to perform the wallpaper display method provided in the first aspect and the various possible implementations of the first aspect.

[0039] In a fifth aspect, the present application provides a computer program product, comprising computer programs / instructions, which, when executed by a processor, implement the wallpaper display method provided in the first aspect and the various possible implementations of the first aspect.

[0040] The advantages of the second aspect to the fifth aspect above can be referred to the related description of the first aspect and the various possible implementations of the first aspect, which will not be repeated here. BRIEF DESCRIPTION OF DRAWINGS

[0041] FIG. 1 shows a lock screen interface for displaying a wallpaper according to an embodiment of the present application.

[0042] FIG. 2a shows a personalized wallpaper display scenario based on a lock screen interface according to an embodiment of the present application.

[0043] FIG. 2b shows a personalized wallpaper display scenario based on a desktop according to an embodiment of the present application.

[0044] FIG. 3 shows a structure of a personalized wallpaper generation model according to an embodiment of the present application.

[0045] FIG. 4a shows a multi-device interaction scenario applicable to a wallpaper display method according to Embodiment 1 of the present application.

[0046] FIG. 4b shows an implementation flow of a wallpaper display method according to Embodiment 1 of the present application.

[0047] FIG. 5a shows a desktop of a mobile phone entering a wallpaper setting interface in response to a user operation according to Embodiment 1 of the present application.

[0048] FIG. 5b shows a setting interface of a mobile phone entering a wallpaper setting interface in response to a user operation according to Embodiment 1 of the present application.

[0049] FIG. 5c shows a desktop and a personalized setting interface of a mobile phone entering a wallpaper setting interface in response to a user operation according to Embodiment 1 of the present application.

[0050] Fig. 5d shows a wallpaper setting interface provided by the embodiment 1 of the present application for a user to specify a wallpaper material image.

[0051] Fig. 5e shows a gallery interface provided by the embodiment 1 of the present application for a user to specify a wallpaper material image.

[0052] Fig. 5f shows a wallpaper preview interface provided by the embodiment 1 of the present application based on an initial generation of a wallpaper material image specified by a user.

[0053] Fig. 6a shows a wallpaper preview interface provided by the embodiment 1 of the present application for displaying a personalized target wallpaper image.

[0054] Fig. 6b shows a desktop interface provided by the embodiment 1 of the present application for displaying a personalized target wallpaper image.

[0055] Fig. 7 shows an implementation flowchart of a wallpaper display method provided by the embodiment of the present application for realizing commodity recommendation.

[0056] Fig. 8 shows a complete implementation process of a wallpaper display method provided by the embodiment of the present application.

[0057] Fig. 9 shows an implementation flowchart of a wallpaper display method provided by the embodiment 2 of the present application.

[0058] Fig. 10 shows a setting interface provided by the embodiment 2 of the present application.

[0059] Fig. 11 shows an implementation flowchart of a wallpaper display method provided by the embodiment 3 of the present application.

[0060] Fig. 12a shows an implementation process of determining clothing control information provided by the embodiment of the present application.

[0061] Fig. 12b shows an implementation process of determining posture control information provided by the embodiment of the present application.

[0062] Fig. 12c shows an implementation process of determining background control information provided by the embodiment of the present application.

[0063] Fig. 12d shows an implementation process of loading style parameters to configure a diffusion model provided by the embodiment of the present application.

[0064] Fig. 13 shows a structure of a mobile phone provided by the embodiment of the present application.

[0065] Fig. 14 shows a structure of a server provided by the embodiment of the present application. DETAILED DESCRIPTION

[0066] In order to make the purposes, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0067] In order to facilitate those skilled in the art to understand the solutions in the embodiments of the present application, some concepts and terms related to the embodiments of the present application will be explained first.

[0068] (1) Artificial intelligence generated content (AIGC): It is a method of generating content based on artificial intelligence technology. It uses artificial intelligence models to automatically generate various types of text, images, audio, video and other content according to given topics, keywords, formats, styles and other conditions. In the related technical solutions in the field of image generation, AIGC technology can specifically use diffusion models to generate images corresponding to prompts based on input user-provided prompt information for guiding the model to generate content.

[0069] (2) Diffusion model: Diffusion model is a mathematical model used to describe the diffusion process in a system, which can be used to describe the diffusion motion between molecules, or the spread of information, ideas or behaviors among people in sociology. In physics, gas molecules diffuse from high concentration areas to low concentration areas, which is similar to the loss of information due to noise interference. Therefore, in the specific field of image generation, diffusion models can learn the information decay caused by noise and then use the learned pattern to generate images.

[0070] It can be understood that the terminal device in the embodiments of the present application can also be referred to as a terminal, a user equipment (UE), a mobile station (MS), a mobile terminal (MT), etc. The terminal device can be a mobile phone, a smart TV, a wearable device, a tablet computer (Pad), a computer with wireless transceiver function, a virtual reality (VR) terminal device, an augmented reality (AR) terminal device, a wireless terminal in a smart city, a wireless terminal in a smart home, etc.

[0071] It should be noted that the steps in the methods and processes of the embodiments of the present application are numbered for ease of reference, and do not limit the sequence, and the sequence between the steps is subject to the literal description.

[0072] As described above, the current wallpaper display scheme cannot meet the user's more personalized customization needs.

[0073] For example, referring to the lock screen interface 101 shown in FIG. 1, the mobile phone 10 only supports the user to set his / her own personal image 102 as the personalized wallpaper of the lock screen interface 101 or the interface of the desktop, etc., to show personality or express emotion. However, the mobile phone 10 cannot support the user to change clothes or change background for the personal image 102, for example, to "wear" a hat or a watch for himself / herself in the personal image 102, or to change a suit for himself / herself, or to change a background of flowers for himself / herself, and so on.

[0074] To solve the above problems, the present application provides a wallpaper display method, which can make the terminal device such as a mobile phone meet the user's needs for personalized customization of the wallpaper of the lock screen interface or the desktop, etc. in multiple dimensions. Specifically, the method adds multiple dimensions of wallpaper elements such as clothes, makeup, wearable devices, etc. to the foreground elements contained in the wallpaper material image specified by the user, adjusts the posture or changes the background elements, etc., and finally processes to obtain a personalized target wallpaper and displays it. The above posture and background elements can also be referred to as wallpaper elements in different dimensions. In the above generated target wallpaper, the foreground elements can be matched with the clothes, makeup, wearable devices, etc. that the user likes, and the foreground elements can also adjust the body posture and can be integrated with the background elements that the user likes. In this way, the user's personalized customization and use requirements for the wallpaper can be met.

[0075] Among them, the wallpaper material image specified by the user can also be described as a first image in some embodiments, which can be the source of the generation of the wallpaper in response to the user's specified operation. In some embodiments, the wallpaper material image can be obtained by the user using the mobile phone 10 to take a photo, or downloaded or saved through the third-party application running on the mobile phone 10, for example, the user can download or save the image of his / her heart when browsing the related interface of the shopping application, media application, etc. running on the mobile phone 10, and then specify it as the wallpaper material image in the setting application or the wallpaper application. In other embodiments, the user can also directly operate to set the picture of his / her heart as the wallpaper in some third-party applications, at which time the third-party application can call the components or services that can implement the wallpaper display method provided by the present application, and generate and display the finally generated personalized target wallpaper in response to the user's operation of adding each dimension of the wallpaper element to the picture of his / her heart that is to be set as the wallpaper, without limitation.

[0076] The foreground element can include a character. The posture of the character, such as the stretching direction of the limbs, can be adjusted based on the wallpaper display method provided in the present application. The stretching direction of the limbs of the character can be determined based on the posture map corresponding to the human body skeleton key points. The wallpaper elements in multiple dimensions, such as clothing, posture, background, etc., can be added to the generated personalized wallpaper by using the control information in multiple dimensions, such as clothing, posture, background, etc., during the wallpaper generation stage. In some embodiments, the wallpaper elements in multiple dimensions described above can also include makeup, wearable device, and other dimensions of wallpaper elements, which are not limited herein.

[0077] In other embodiments, the wallpaper elements in multiple dimensions described above can also include modification elements of the foreground elements, such as text, stickers, heart-shaped bubbles, and other shape bubbles, which are not limited herein.

[0078] It can be understood that the control information in multiple dimensions described above can be generated based on system default recommendations or user specifications, and can be used as input data of the preset diffusion model for generating personalized wallpaper. At the same time, the diffusion model can load the default recommended or user specified style parameters to configure the corresponding style parameters according to the user specified wallpaper material image. Further, the configured diffusion model can use the control information in multiple dimensions described above to add clothing, background, and other elements to one or more target characters in the wallpaper material image, and adjust the posture of the characters, etc., to finally generate a personalized wallpaper that meets the user's individual customization needs. The user only needs to determine to set the generated personalized wallpaper as the wallpaper of the lock screen interface or the interface of the desktop of the terminal device such as a mobile phone, so that the terminal device such as a mobile phone displays the personalized wallpaper when the conditions for displaying the wallpaper are met.

[0079] It can be understood that, taking a character as an example, the foreground element can be obtained based on the face control information extracted from the user specified wallpaper material image. Correspondingly, the foreground element can include face control information of one or more target characters. The added control information in multiple dimensions, such as clothing, posture, background, etc., can be preset on the server or the terminal device such as a mobile phone, which is used to control the generation effect of the personalized wallpaper during the process of generating the personalized wallpaper according to the user specified wallpaper material image. In other embodiments, the added control information in multiple dimensions, such as clothing, posture, background, etc., can also be generated according to the corresponding specified related material image of the user, which is not limited herein.

[0080] As an example, the above-mentioned clothing control information can include a feature vector or a set of feature vectors corresponding to a hat, a piece of clothing, a pair of shoes, a necklace, a bracelet, an earring, and the like. The above-mentioned posture control information can include a feature vector or a set of feature vectors for describing various postures such as standing, running, jumping, bending, and the like. The above-mentioned background control information can include a feature vector or a set of feature vectors corresponding to natural scenery elements such as flowers, mountains and rivers, trees and forests as backgrounds, and can also include a feature vector or a set of feature vectors corresponding to human landscape elements such as houses, buildings, historical sites, and the like as backgrounds. The background control information can be provided by a corresponding background template, or can be obtained by extracting from a user-specified background element image, which is not limited herein.

[0081] In some embodiments, the above-mentioned control information of multiple dimensions can also include makeup control information corresponding to a makeup, and device control information corresponding to a wearable device, and the like, wherein the device control information can include, for example, a feature vector or a set of feature vectors corresponding to a watch, a headset, glasses, and the like. Correspondingly, the makeup control information can be provided by a corresponding makeup template, or can be obtained by extracting from a user-specified makeup element image, which is not limited herein. The device control information can be provided by a corresponding device template, or can be obtained by extracting from a user-specified device element image, which is not limited herein. The following continues to take the wallpaper elements and related control information of the dimensions of clothing, posture, background, and the like as an example to introduce the specific implementation process of the wallpaper display method provided by the present application.

[0082] It can be understood that the above-mentioned control information of multiple dimensions such as face, clothing, posture, background, and the like can be a multi-dimensional feature vector or other types that can be used as input data of the above-mentioned diffusion model, which is not limited herein. Among them, the above-mentioned diffusion model for generating personalized wallpaper can be a pre-trained deep learning model or neural network model capable of generating personalized wallpaper based on wallpaper material images and input control information, and the like. The process of generating personalized wallpaper using the diffusion model, i.e., the process of generating and displaying personalized wallpaper by AIGC method, can be implemented in detail by referring to the detailed description below in conjunction with the related drawings, which is not described herein. The following continues to take the control information of multiple dimensions such as face, clothing, posture, background, and the like as an example to introduce the specific implementation process of the wallpaper display method for generating personalized wallpaper provided by the present application.

[0083] Correspondingly, the above-mentioned loadable style parameters are used to guide the diffusion model to generate personalized wallpaper of a corresponding style. Among them, the above-mentioned style can include but is not limited to futuristic technology style, dream style, ancient style, retro style, modern style, fashion magazine style, college style, pastoral style, oil painting style, and the like.

[0084] It can be understood that the above loadable style parameters can be matched according to the control information of the above multiple dimensions of face, clothing, posture, background, etc., or can be loaded according to the wallpaper style specified by the user, which is not limited here.

[0085] It can be understood that the above processing obtains a personalized target wallpaper, which can be simply referred to as a personalized wallpaper or a target wallpaper in the embodiments of the present application, and in some embodiments, it can also be described as a second image, wherein the second image is generated based on the first image and the input operation of the user. The input operation can not only specify the first image, i.e., the wallpaper material image, as the generation source of the wallpaper, but also specify at least one of the clothing, makeup, wearable device, background, and style of the person in the personalized wallpaper or target wallpaper obtained by processing.

[0086] In some embodiments, the wallpaper elements in the personalized target wallpaper obtained by the above processing, such as clothing, makeup, wearable devices, etc., which are matched with the foreground elements, can also be associated with relevant commodity link recommendations for the user to purchase. As an example, for clothing such as hats, clothes, shoes, necklaces, bracelets, earrings, etc., the commodity link recommended by the shopping application or browser application can be correspondingly associated. For the use effect of makeup such as eyebrow pencil, eye shadow, lipstick, etc., and the use effect of skin care products such as toner, emulsion, etc., the commodity link of the relevant makeup and skin care products recommended by the shopping application or browser application can be correspondingly associated. For wearable devices such as watches, earphones, glasses, etc., the commodity link recommended by the shopping application or browser application can be correspondingly associated. For the replaced background elements such as natural scenery such as flowers, rivers and mountains, trees and forests, etc., the commodity link recommended by the tourism application or browser application can be correspondingly associated, which corresponds to the introduction and purchase channel of the tourism service products that can be provided. When the user clicks on the relevant commodity link in the wallpaper displayed on the terminal device, the terminal device such as a mobile phone can display the relevant commodity shopping interface for the user to quickly purchase. In this way, the wallpaper display method provided by the present application can recommend the desired commodities for the user through the personalized wallpaper, provide more convenient services for the user, and be beneficial to improve the user's experience.

[0087] In addition, the personalized target wallpaper obtained by the above processing can also be recommended to a third-party application, for example, recommended to a shopping application as a buyer show, or recommended to some short video application or other media application as a material for publishing text and pictures, etc., which is not limited here.

[0088] As an example, FIG. 2a illustrates a personalized wallpaper display scene according to an embodiment of the present application.

[0089] As shown in FIG. 2a, the lock screen interface 210 displayed by the mobile phone 10 can include a personalized wallpaper 211 set by the user. The personalized wallpaper 211 can include a target person 212, and dress elements such as a hat 213 and a watch 214 added to the target person, and elements such as a flower background 215 added to the target person. Among them, the facial features of the target person 212 can be obtained based on the user-specified wallpaper material image extraction. In other embodiments, the posture of the target person 212 in the personalized wallpaper 211 displayed by the mobile phone 10 can also be different from the posture of the person in the user-specified wallpaper material image. Moreover, in some embodiments, if the style parameter corresponding to the personalized wallpaper 211 described above is a pastoral style, the personalized wallpaper 211 described above can also present a color effect of the pastoral style.

[0090] In this way, the personalized wallpaper 211 shown in FIG. 2a described above can meet the user's individual customization needs for dress, posture, and background, etc.

[0091] Similarly, FIG. 2b illustrates another personalized wallpaper display scenario according to an embodiment of the present application.

[0092] As shown in FIG. 2b, the desktop 220 displayed by the mobile phone 10 can include a personalized wallpaper 221 set by the user. The personalized wallpaper 221 can also include a face of a target person, and dress elements such as a hat and a watch added to the target person, and elements such as a flower background added to the target person. It can be understood that on the desktop 220, part of the content of the personalized wallpaper 221 described above can be obscured by service cards and application icons, and in the case that the mobile phone 10 displays different main interfaces of the desktop 220, the part of the content of the personalized wallpaper 221 described above that is obscured can be different. Correspondingly, when the user operates the mobile phone 10 on the desktop 220, no matter which main interface the desktop 220 switches to, the service cards, application icons, and other elements on each main interface can be different, and the user can see the personalized wallpaper as the desktop background, so as to meet the user's individual customization needs for the wallpaper, thus improving the user's visual experience.

[0093] Further, based on the wallpaper display method provided in the present application, each wallpaper element in the personalized wallpaper displayed by the terminal device such as the mobile phone can also be bound with an associated commodity link, and then the corresponding commodity link can be pushed to the user through the personalized wallpaper, improving the advertising value of the personalized wallpaper. Among them, the commodity link can include purchase links of the same or similar goods provided by various shopping malls or shopping applications, and can also include purchase links of tourism service products provided by various tourism applications, etc., which are not limited here.

[0094] In some embodiments, the wallpaper element in the personalized wallpaper with the commodity link bound thereto can display a more eye-catching mark, such as the search mark 216 marked on the hat 213 in FIG. 2a. In other embodiments, the mark can also be in other mark styles different from the search mark 216 in FIG. 2a, such as a "+" mark, a "·" mark, or a dynamically flashing mark, and the like, which are not limited herein.

[0095] In some optional embodiments, the mark for marking the commodity link can be added to the wallpaper element in the corresponding personalized wallpaper when it is detected that the user shows interest in the personalized wallpaper displayed by the terminal device such as a mobile phone. The behavior of the user showing interest in the personalized wallpaper can include, but is not limited to, the operation behavior of the user clicking or circling at least one wallpaper element in the personalized wallpaper, or the operation behavior of the user operating to collect the generated personalized wallpaper image, the operation behavior of the user taking a screenshot, locally magnifying, saving, or further editing the personalized wallpaper displayed on the lock screen interface or the desktop, and the like, which are not limited herein.

[0096] It can be understood that the image processing process of the wallpaper material image involved in the wallpaper display method provided in the present application can be implemented by a personalized wallpaper generation model deployed on a server, which can include the diffusion model, the database capable of providing the control information in the multiple dimensions, and the control network, and the like. The server obtains the personalized wallpaper meeting the requirements by using the wallpaper display method provided in the present application to process the wallpaper material image including one or more target persons provided by the terminal device such as a mobile phone, and then sends the personalized wallpaper to the terminal device such as a mobile phone as a wallpaper preview image or a wallpaper to present to the user. The scenario of implementing the wallpaper display method provided in the present application through the interaction between the terminal device such as a mobile phone and the server can be referred to as a multi-device interaction scenario. It can be understood that the terminal device interacting with the server in the multi-device interaction scenario can include one or more, such as the scenario of the interaction between a mobile phone and a server, or the scenario of the interaction between a mobile phone and a server, respectively, and the like, which are not limited herein.

[0097] The diffusion model for generating personalized wallpaper described above, together with other algorithms or models for extracting target characters in wallpaper material images, and algorithms or models capable of providing dimensional control information or style parameters, etc., can constitute an algorithm library or model set for generating personalized wallpaper, which can be collectively referred to as a personalized wallpaper generation algorithm or a personalized wallpaper generation model hereinafter. In other embodiments, the personalized wallpaper generation model can also be pre-deployed on a terminal device such as a mobile phone to perform image processing on a wallpaper material image specified by user operation, obtain a personalized wallpaper, and display. The terminal device such as a mobile phone that has deployed the personalized wallpaper generation model can have strong computing power. In this scenario, the terminal device such as a mobile phone can implement the wallpaper display method provided by the present application as a separate execution subject. Therefore, the scenario in which the terminal device such as a mobile phone implements the wallpaper display method provided by the present application alone can be referred to as a single-device implementation scenario.

[0098] As an example, FIG. 3 illustrates a structural diagram of a personalized wallpaper generation model according to an embodiment of the present application.

[0099] It can be understood that the personalized wallpaper generation model deployed on the server or deployed on the terminal device such as a mobile phone can have the model structure or model composition shown in the example of FIG. 3. In other embodiments, the personalized wallpaper generation model described above can also include more or fewer components, or can be implemented by a pre-trained neural network model, which is not limited here.

[0100] As shown in FIG. 3, the personalized wallpaper generation model 300 can include a portrait analysis module 310, a face control module 321, a clothing control module 322, a posture control module 323, a background control module 324, a style control module 325, and a diffusion model 330. The face control module 321, the clothing control module 322, the posture control module 323, the background control module 324, the style control module 325, etc. can all be control models implemented based on a control network, and can provide corresponding dimensional control information as input information of the diffusion model 330.

[0101] The portrait analysis module 310 is configured to perform face detection, age detection, gender analysis, etc. on the target person in the "wallpaper material image" specified by the user. The face detection result can be provided to the face control module 321 to extract face features. The face control module 321 can input the extracted face feature information as face control information of the target person to the diffusion model 330 to control the face of the person foreground element in the finally generated personalized wallpaper to be the same or similar to the face of the target person. The age, gender, etc. determined by the portrait analysis module 310 can be used as a basis for other control modules to determine the wallpaper elements of the recommended personalized wallpaper, for example, to determine the clothing elements for the clothing control module 322, to determine the pose adjustment result of the target person for the pose control module 323, and to determine the background elements for the background control module 324, etc.

[0102] The clothing control module 322 is configured to provide clothing control information for the diffusion model 330 to add clothing elements as wallpaper elements for the target person when generating a personalized wallpaper. The specific implementation process of the clothing control information provided by the clothing control module 322 will be described in detail below in conjunction with related embodiments and drawings, and will not be described here.

[0103] The pose control module 323 is configured to provide pose control information for the diffusion model 330 to add pose elements as wallpaper elements for the target person when generating a personalized wallpaper. The specific implementation process of the pose control information provided by the pose control module 323 will be described in detail below in conjunction with related embodiments and drawings, and will not be described here.

[0104] The background control module 324 is configured to provide background control information for the diffusion model 330 to add background elements as wallpaper elements for the target person when generating a personalized wallpaper. The specific implementation process of the background control information provided by the background control module 324 will be described in detail below in conjunction with related embodiments and drawings, and will not be described here.

[0105] The style control module 325 is configured to provide style control information. In the embodiments of the present application, the style control module 325 can provide style parameter configurations of the diffusion model corresponding to different wallpaper styles through the stylization model to realize control of the wallpaper style. In other embodiments, the style control module 325 can also generate prompt information corresponding to different wallpaper styles through the prompt generation model as input of the diffusion model to realize control of the wallpaper style. Based on this, in some embodiments of the present application, the style control module 325 can also be referred to as a stylization model library.

[0106] It is understood that the aforementioned face control module 321, clothing control module 322, posture control module 323, and background control module 324 can all provide control information as part of the input data for the diffusion model 330. The style control module 325 can provide corresponding style parameters for configuring the model's related parameters. The face control information, clothing control information, posture control information, and background control information provided by the face control module 321, clothing control module 322, posture control module 323, and background control module 324 can be used as input data for the diffusion model 330 after configuring the style parameters.

[0107] The diffusion model 330 can generate a personalized target wallpaper image based on the wallpaper material image specified by the user and the control information corresponding to the wallpaper elements in each dimension. In this embodiment, it can also be referred to as a personalized wallpaper.

[0108] Based on the personalized wallpaper display scenarios shown in Figures 2a and 2b above and the structure of the personalized wallpaper generation model shown in Figure 3, the following describes in detail the specific implementation process of the wallpaper display method provided in this application in different interactive scenarios, combined with different embodiments.

[0109] The following section, in conjunction with Embodiment 1, describes a specific implementation process of the wallpaper display method provided in this application in a multi-device interaction scenario.

[0110] Example 1

[0111] Taking mobile phone 10 and server 20 as examples, Figure 4a shows a schematic diagram of a multi-device interaction scenario to which a wallpaper display method is applicable, according to an embodiment of this application.

[0112] As shown in Figure 4a, the multi-device interaction scenario may include a mobile phone 10 and a server 20. The mobile phone 10 is only one example of a terminal device. In other embodiments, the multi-device interaction scenario may also include interaction scenarios between other terminal devices and the server, such as interaction between a tablet computer and the server, or interaction between a laptop computer and the server, etc., without limitation.

[0113] Taking the iPhone 10 as an example, the iPhone 10 can provide a user interface (UI), such as an interface for users to operate and set wallpapers, to receive user input. This input may include operations such as the user specifying wallpaper images containing one or more of the aforementioned target characters, etc. For details, please refer to the relevant interface descriptions below, which will not be elaborated here.

[0114] Referring again to Figure 4a, server 20 may have a network transceiver 210 for maintaining communication with terminal devices such as mobile phone 10. As mentioned above, the personalized wallpaper generation model described above can be deployed on server 20. This model can be executed by one or more processors on server 20, such as a neural processing unit (NPU) 220 (hereinafter referred to as NPU processor 220) or a central processing unit (CPU) 230 (hereinafter referred to as CPU processor 230) on server 20, thereby implementing the wallpaper display method provided in this application.

[0115] Based on the multi-device interaction scenario shown in Figure 4a above, the interaction process of the wallpaper display method provided in this application will be described in detail below with reference to the relevant accompanying drawings.

[0116] As mentioned earlier, in the specific implementation of the wallpaper display method provided in this application, the wallpaper elements added to the generated personalized wallpaper can be recommended by the system by default or specified by the user, and the style of the generated personalized wallpaper can also be recommended by the system by default or specified by the user. The following section, referring to the interaction flow shown in Figure 4b, introduces the specific implementation process of a wallpaper display method that adds wallpaper elements based on default recommendations, controls the wallpaper style to generate and display personalized wallpapers.

[0117] Figure 4b illustrates a schematic diagram of the implementation process of a wallpaper display method according to an embodiment of this application. It can be understood that in this embodiment, the implementation process may involve the interaction between the mobile phone 10 and the server 20. The server 20 can be a wallpaper server or a theme server, and a personalized wallpaper generation model for generating personalized wallpapers is pre-deployed on the server 20. This model may include a diffusion model, a control model for providing a set of feature vectors related to wallpaper elements such as faces, clothing, poses, and backgrounds (an example of control information), and style parameters or prompts corresponding to various different styles, etc., which are not limited here.

[0118] Specifically, as shown in Figure 4b, the implementation process may include the following steps:

[0119] S401: Phone 10 detected the user's action of setting the wallpaper and displayed the relevant settings interface.

[0120] For example, users can set wallpapers on the UI of terminal devices such as mobile phones 10, including setting system-recommended wallpapers, selecting images from the gallery as wallpapers, and setting personalized wallpapers, etc. In this embodiment, when mobile phone 10 detects the user's operation of specifying a personalized wallpaper, it can display the relevant settings interface for customizable personalized wallpapers.

[0121] As an example, Figures 5a to 5c show schematic diagrams of the relevant interfaces of a mobile phone responding to user operations to enter the wallpaper settings interface.

[0122] As shown in Figure 5a, the user can click the settings application icon 511 on the desktop 510 displayed on the phone 10 to display the settings interface 520 shown in Figure 5b.

[0123] As shown in Figure 5b, the user can continue to click the "Desktop and Personalization" function 521 on the settings interface 520 displayed on the phone 10, so that the phone 10 displays the desktop and personalization settings interface 530 shown in Figure 5c.

[0124] As shown in Figure 5c, if the user continues to click the wallpaper option 531 on the desktop and personalization settings interface 530 displayed on the mobile phone 10, the mobile phone 10 will display the wallpaper settings interface 540 shown in Figure 5d, and then the user can set a wallpaper. It can be understood that setting a wallpaper on this wallpaper settings interface 540 can include setting a personalized wallpaper, which can be obtained based on the wallpaper display method provided in this application. Specifically, the process by which the mobile phone 10 responds to the user's wallpaper setting operation, obtains and displays the personalized wallpaper can be further referred to in the relevant descriptions of steps S402 to S413 below, and will not be repeated here.

[0125] S402: Mobile phone 10 obtains the wallpaper image specified by the user in the relevant settings interface.

[0126] For example, the wallpaper image specified by the user can be used as image data input to a pre-deployed diffusion model 330 or an AIGC-based portrait analysis module 310 for analyzing the facial features, age, gender, and posture of individuals in the image. The wallpaper image may include people. In this embodiment, taking people as an example, the wallpaper image specified by the user may include facial control information of one or more people, hereinafter referred to as the target person.

[0127] In some embodiments, the aforementioned settings interface may be a related interface provided by a settings application or wallpaper application. In other embodiments, the aforementioned settings interface may also be a related interface provided by a shopping application or media application browsed by the user, and no limitation is made here.

[0128] As an example, Figures 5d to 5f show schematic diagrams of the interface for a mobile phone to obtain wallpaper material images and customize personalized wallpapers in response to user operations.

[0129] As shown in Figure 5d, on the wallpaper settings interface 540, the user can click the option 541 corresponding to "Select from Gallery" to select a photo with a person as the specified wallpaper material image. As mentioned earlier, in this embodiment, the wallpaper material image may include one or more people, which can support the subsequent extraction of one or more target people by the portrait analysis module 310.

[0130] As shown in Figure 5e, in response to the user clicking the "Select from Gallery" option 541, the mobile phone 10 can display the gallery interface 550. The user can select a favorite photo as wallpaper image on the gallery interface 550, such as a portrait photo 551. Correspondingly, the mobile phone 10 can display the wallpaper preview interface 560 shown in Figure 5f.

[0131] In this embodiment, the mobile phone 10 can display a "personalization" control 561 on the displayed wallpaper preview interface 560. The user can click the "personalization" control 561 to combine a specified personal portrait 551 with one or more other wallpaper elements to generate a personalized wallpaper. In other embodiments, the mobile phone 10 and other terminal devices can also display controls supporting customized personalized wallpapers on interfaces different from the wallpaper preview interface 560 illustrated in Figure 5f, or display other forms of "personalization" controls on the wallpaper preview interface 560 illustrated in Figure 5f; this is not limited here. Furthermore, the display location of the "personalization" control or the application that triggers the display of the "personalization" control is not specifically limited; for example, a theme application can provide related functional controls containing the aforementioned "personalization" control through the personalized wallpaper function.

[0132] S403: Mobile phone 10 sends wallpaper material images to server 20.

[0133] For example, in this embodiment of the application, the personalized wallpaper generation model described above can be deployed on server 20. Therefore, after obtaining the wallpaper material image specified by the user, such as the personal portrait photo 551 mentioned above, mobile phone 10 can send the wallpaper material image to server 20.

[0134] S404: Server 20 performs portrait analysis on each target person in the wallpaper material image and extracts facial features as facial control information.

[0135] For example, after the server 20 obtains the wallpaper image sent by the mobile phone 10, it can use the wallpaper image as input data for the portrait analysis module 310 in the personalized wallpaper generation model. This allows for face detection, age detection, and gender analysis of the wallpaper image. The extracted facial features corresponding to the face detection results can be used as input information for the face control module 321 in the personalized wallpaper generation model. Furthermore, the face control module 321 can generate face control information for each target person based on the input facial features. The wallpaper image can include one or more target people; correspondingly, extracting facial features from the wallpaper image can generate face control information for each of the one or more target people.

[0136] It is understandable that the aforementioned facial control information can be the unique identification information (ID) of each target person, and therefore can also be called the target person's facial ID.

[0137] S405: Server 20 retrieves the default recommended clothing control information.

[0138] For example, server 20 may have a preset feature vector library and / or outfit template library containing various clothing elements. During the process of obtaining personalized wallpaper based on the aforementioned wallpaper material image processing, server 20 can obtain feature vectors or feature vector sets of one or more clothing elements from the feature vector library as clothing control information for the target person, or clothing matching recommendation results. Server 20 can also obtain any recommended outfit template from the aforementioned outfit template library, and use the feature vectors or feature vector sets corresponding to the outfit template that describe each clothing element as clothing control information for the target person. Correspondingly, server 20 can use the feature vectors or feature vector sets of one or more clothing elements contained in the recommendation results as default recommended clothing control information. In this embodiment, the algorithm or capability model for obtaining clothing control information from the feature vector library containing various clothing elements can be integrated into the aforementioned personalized wallpaper generation model, serving as the clothing control module 322 providing the aforementioned ability to obtain clothing control information.

[0139] It is understood that the above-mentioned recommended results or default recommended clothing control information can be any one of the default clothing combinations. For example, for the same or different target characters, the clothing combinations recommended by server 20 can be randomly selected from multiple combinations to provide users with different matching effects to choose from.

[0140] As mentioned earlier, the aforementioned clothing control information may include feature vectors corresponding to elements such as hats, clothes, shoes, necklaces, bracelets, and earrings. Correspondingly, the various clothing combinations mentioned above may include any combination of clothing elements such as hats, clothes, shoes, necklaces, bracelets, and earrings. For example, a clothing combination may include a set of clothes and combinations of hats, shoes, and necklaces to match that set. Another example is that a clothing combination may include combinations of tops, pants, shoes, necklaces, bracelets, and earrings in matching colors, and so on.

[0141] Specifically, the process by which server 20 obtains the default recommended clothing control information will be described in detail below with reference to the accompanying drawings, and will not be repeated here.

[0142] S406: Server 20 obtains the default recommended attitude control information.

[0143] For example, a pose template library can be preset on server 20. During the process of obtaining personalized wallpaper based on the aforementioned wallpaper material images, server 20 can obtain any recommended pose template from the pose template library and use the feature vector or set of feature vectors corresponding to the pose template to describe the pose as pose control information for the target person. In this embodiment, the algorithm or capability model for extracting pose control information based on the pose templates provided by the pose template library can be integrated into the aforementioned personalized wallpaper generation model, serving as the pose control module 323 providing the capability to extract the aforementioned pose control information. This pose control information can be used to control the pose maintained by the target person in the generated personalized wallpaper, including but not limited to standing, running, jumping, and bending postures, without limitation.

[0144] It is understood that any of the above-mentioned recommended pose templates can be a pose template randomly selected from multiple pose templates in the pose template library, or a pose template recommended in sequence in different personalized wallpaper generation processes according to a preset recommendation order. There are no restrictions here.

[0145] As mentioned earlier, the preset posture template library on server 20 may include posture templates corresponding to postures such as standing, running, jumping, and bending over. In some embodiments, when the wallpaper material image includes multiple target characters, server 20 can load the same or different posture templates for each target character. The same posture template can control different target characters to maintain the same posture, while different posture templates can control different target characters to maintain different postures. It can be understood that when server 20 loads different posture templates to control the postures maintained by different target characters, the postures corresponding to the loaded posture templates can have a matching relationship. This can avoid the situation where the postures maintained by different target characters in the final generated personalized wallpaper conflict with each other and affect the user experience.

[0146] Specifically, the process by which server 20 obtains the default recommended attitude control information will be described in detail below with reference to the accompanying drawings, and will not be repeated here.

[0147] S407: Server 20 retrieves the default recommended background control information.

[0148] For example, a background control model can be preset on server 20. This background control model can provide the ability to process one or more background elements into a background element image and to convert the background element image into background control information. Each background element can be pre-added to a corresponding element library preset on server 20. In this embodiment, the background control model can be integrated into the personalized wallpaper generation model, serving as background control module 324 to provide the aforementioned ability to process background element images and generate background control information. During the process of obtaining a personalized wallpaper based on the wallpaper material image, server 20 can use the background control model to obtain one or more background elements from the preset element library to generate a background element image, or server 20 can use the background control model to load an existing background element image and convert it into background control information, such as feature vectors or feature vector sets containing one or more background elements.

[0149] It is understood that the aforementioned background control information can be used to control one or more background elements of the final generated personalized wallpaper. These background elements, in conjunction with the target person (as a foreground element) extracted from the wallpaper material image, can complement the foreground element. This background control information can include, but is not limited to, natural scenery such as flowers, mountains, rivers, trees, and forests, as well as cultural landscapes such as buildings and historical sites, etc., and is not limited here.

[0150] It is understood that the background elements recommended by the background control model for the target person may be background elements selected by the background control model from the element library that correspond to the image effects of various background elements, or background elements in background element images randomly loaded by the background control model, or background elements in background element images matched by the background control model for the target person based on the above clothing control information and / or posture control information, etc., and so on. No restrictions are imposed here.

[0151] Specifically, the process by which server 20 obtains the default recommended background control information will be described in detail below with reference to the accompanying drawings, and will not be repeated here.

[0152] It is understood that in other embodiments, during the process of the server 20 generating personalized wallpapers using the wallpaper display method provided in this application, it can obtain more dimensions of control information corresponding to the wallpaper material image specified by the user. For example, it can also obtain makeup control information, wearable device control information, and decorative element control information based on related element images or templates, etc., to add more wallpaper elements to the target person (as a foreground element) in the wallpaper material image, which is not limited here. Among them, wallpaper elements of clothing, makeup, wearable devices, etc., can be merged with the target person and used as the foreground elements of the final generated personalized wallpaper, while wallpaper elements of background and decorative elements (such as text, stickers, hearts, or other shaped bubbles, etc.) can be used as the background elements of the final generated personalized wallpaper to complement or modify the foreground elements. In other embodiments, the above-mentioned decorative elements can also be merged with the target person as a foreground element and used as a foreground element, which is not limited here.

[0153] In some embodiments of this application, the server 20 obtains recommendation rules corresponding to the control information of the above-mentioned multiple dimensions for the target person as the foreground element based on the personalized wallpaper generation model. It can also be determined based on the user's historical behavior data provided by the mobile phone 10. That is, it can intelligently recommend and obtain control information of various dimensions such as clothing, makeup, wearable devices, posture and background that the user may be interested in based on the historical behavior data generated by the user's mobile phone 10, so as to control the final generated personalized wallpaper to better meet the user's personalized customization needs. No limitation is made here.

[0154] In other embodiments, the server 20 may also obtain the corresponding control information for each dimension based on the image uploaded by the user via the mobile phone 10 that can provide the control information for each dimension mentioned above, so as to control the finally generated personalized wallpaper to better meet the user's personalized customization needs. The specific implementation process of this method will be described in detail below in conjunction with Embodiment 2, and will not be repeated here.

[0155] S408: Server 20 loads the style control information corresponding to the default recommended style and configures or inputs the diffusion model. This style control model may include style parameters or the prompt information generated for the default style.

[0156] For example, server 20 can preset one or more style parameters corresponding to different styles. The style control information provided by these style parameters can control the style presented by the final generated personalized wallpaper, including but not limited to futuristic, dreamy, ancient, retro, modern, fashion magazine, academic, pastoral, and oil painting styles. The style parameters corresponding to different styles can be pre-trained. After loading a default recommended style parameter, server 20 can obtain that style parameter and configure the diffusion model 330. The algorithm or functional model for loading the style parameter can be pre-integrated into the personalized wallpaper generation model, for example, as a style control module 325 providing the ability to obtain the style parameter and configure the diffusion model 330.

[0157] In other embodiments, server 20 may also preset multiple styles and a prompt generation model capable of organizing text according to each style to generate corresponding prompt information. This prompt generation model can be an algorithm or neural network model based on the aforementioned AIGC technology to generate prompt information describing the corresponding style. After training, this prompt generation model can generate corresponding prompt information for different wallpaper styles. For example, based on the default recommended ancient style, the prompt generation model can generate prompt information describing the color scheme, decorative elements, etc., of this ancient style, which will not be elaborated further here.

[0158] It is understandable that the style parameters corresponding to the default recommended style loaded by server 20 can be the result of random recommendations from a variety of preset style parameters, or the result of recommendations in a predetermined order during different personalized wallpaper generation processes. Similarly, server 20 can generate corresponding prompt information based on the default recommended style, or it can randomly recommend a style from a variety of preset styles and load the prompt generation model to generate the corresponding prompt information; no restrictions are imposed here.

[0159] Specifically, the implementation process of server 20 loading the default recommended style parameter configuration diffusion model or loading the prompt generation model to generate the prompt information corresponding to the default recommended style will be described in detail below with reference to the relevant figures, and will not be repeated here.

[0160] S409: Server 20 inputs face control information, clothing control information, posture control information, and background control information into the configured diffusion model to generate a personalized target wallpaper image.

[0161] For example, the server 20 can input the clothing control information, posture control information, and background control information obtained in the above steps S405 to S407 into the diffusion model 330 configured in the above step S408, so as to control the generation of personalized wallpaper, that is, personalized target wallpaper image.

[0162] It is understood that after the style parameters are configured, the above-mentioned diffusion model 330 can extract the face of the target person corresponding to the above-mentioned wallpaper material image, and based on the corresponding face control information and the above-mentioned clothing, posture, background and other control information as input information, control the generation of personalized wallpaper images with a specific style and multiple dimensions of wallpaper elements such as the above-mentioned face, clothing, posture, background and other elements. The specific generation process will be described in detail below with reference to the relevant figures, and will not be repeated here.

[0163] S410: Server 20 sends the target wallpaper image to mobile phone 10.

[0164] For example, after generating a personalized target wallpaper image, server 20 can return the target wallpaper image to mobile phone 10. As mentioned above, the foreground elements in the target wallpaper image can be target figures extracted from wallpaper material images sent by mobile phone 10, as well as wallpaper elements such as clothing, makeup, and wearable devices added by server 20 to each target figure based on the aforementioned personalized wallpaper generation model, without limitation. Furthermore, the background elements of the target wallpaper image can be background element images recommended and configured by server 20 for the foreground elements based on the aforementioned personalized wallpaper generation model, and can also include background element images or decorative elements added to the foreground elements, without limitation.

[0165] S411: Phone 10 displays the target wallpaper preview interface.

[0166] For example, after obtaining the personalized target wallpaper image returned by the server 20, the mobile phone 10 can display the target wallpaper image on the wallpaper preview interface.

[0167] As an example, Figure 6a shows a schematic diagram of a wallpaper preview interface according to an embodiment of this application.

[0168] As shown in Figure 6a, the wallpaper preview interface 610 displayed on the mobile phone 10 may include the target wallpaper image 611 generated and returned on the server 20 as described above. In some embodiments, the wallpaper preview interface 610 may display a lock screen interface with the target wallpaper image set as the lock screen wallpaper. In other embodiments, it may also present other interface styles, which are not limited here.

[0169] Referring again to Figure 6a, the wallpaper preview interface 610 may also include a "Personalization" control 612, a "Blur" control 613, and an application control 614. The "Personalization" control 612 can trigger the regeneration of a personalized wallpaper. The functions supported by this control for generating personalized wallpapers are the same as those of the "Personalization" control 561 shown in Figure 5f above, and will not be elaborated upon here. The "Blur" control 613 allows users to further process the currently previewed target wallpaper image 611, such as blurring the background. The application control 614 allows users to set the target wallpaper image displayed in the current wallpaper preview interface 610 as the lock screen wallpaper or the desktop wallpaper, etc.

[0170] S412: Phone 10 responds to a user's action of setting a target wallpaper image as the wallpaper and completes the wallpaper setting.

[0171] For example, the mobile phone 10 can respond to the user's operation of clicking the application control 614 in the wallpaper preview interface 610 shown in Figure 6a above, and set the target wallpaper image as the wallpaper of the lock screen and / or desktop to complete the wallpaper operation.

[0172] S413: Phone 10 responds to user actions that instruct it to display the lock screen or desktop, and displays the target wallpaper.

[0173] For example, after setting the wallpaper, the mobile phone 10 can jump to the desktop, such as desktop 620 as shown in Figure 6b, whose wallpaper can correspond to the target wallpaper image set above. When the mobile phone 10 detects the user's lock screen operation, the mobile phone 10 can also display the corresponding lock screen interface. The desktop or lock screen interface displayed by the mobile phone 10 can include the personalized target wallpaper image mentioned above. The desktop or lock screen interface can also refer to Figure 2b or Figure 2a and related descriptions, respectively, and will not be elaborated upon here. In other embodiments, after setting the wallpaper, the mobile phone 10 can also display the relevant interface shown in Figure 5c or Figure 5d, or other relevant interfaces that facilitate other settings by the user, without limitation.

[0174] Based on the interactive implementation process of steps S401 to S413 above, the wallpaper display method provided by this application can provide personalized wallpaper display function for terminal devices such as mobile phones 10. For example, it calls the server 20 to match multiple dimensions of wallpaper elements in the foreground elements of the wallpaper material image specified by the user, and finally obtains and displays a personalized wallpaper with diversified wallpaper elements, which the user can choose to set as the lock screen interface or desktop wallpaper, etc., thus meeting the user's needs for customization or use of personalized wallpapers.

[0175] As mentioned above, based on the wallpaper display method provided in this application, the personalized wallpaper displayed by mobile phone 10 can recommend purchase links for desired products to the user. Referring to Figure 2a above, in the lock screen interface displayed by mobile phone 10, some wallpaper elements of the personalized target wallpaper image can display indicators indicating that they are associated with related products, such as the search mark 216 on the hat shown in Figure 2a. When mobile phone 10 detects that the user clicks on the indicator on the wallpaper element, mobile phone 10 can display related products recommended to the user. These related products can be the same as or highly similar to the corresponding wallpaper element, and there are no restrictions on this.

[0176] In this embodiment, the mobile phone 10 can add the aforementioned product identifiers to the target wallpaper image displayed on the lock screen or desktop. Specifically, the mobile phone 10 can predict whether the user is interested in the target wallpaper image based on user behaviors such as collecting, screenshotting, lingering, returning to view, and saving. If the user is predicted to be interested in the target wallpaper image, the phone can identify wallpaper elements that the user is more interested in within the target wallpaper image, search for identical or similar products in the installed shopping applications, and then add product identifiers to the relevant wallpaper elements of the target wallpaper image and associate them with the searched product purchase links.

[0177] As an example, Figure 7 illustrates a schematic diagram of the implementation process of a wallpaper display method for product recommendation according to an embodiment of this application.

[0178] It is understood that the entities executing each step in this implementation process can be terminal devices such as mobile phones. To simplify the description, the entities executing each step will not be repeated when describing the specific steps in the following implementation process.

[0179] Specifically, as shown in Figure 7, the implementation process of this wallpaper display method may include the following steps:

[0180] S701: Display the target wallpaper image.

[0181] For example, the mobile phone 10 can display the target wallpaper image on the lock screen, desktop, or wallpaper preview screen, which can also be referred to as the target wallpaper.

[0182] As mentioned earlier, the wallpaper elements in the target wallpaper image can include elements from multiple dimensions such as clothing, makeup, wearable devices, and backgrounds. Correspondingly, each wallpaper element can also be associated with products from various dimensions such as clothing, makeup, wearable devices, and tourist attractions. For details on the identification process of each wallpaper element and its corresponding product, please refer to the description of the relevant steps below; these will not be elaborated upon here.

[0183] S702: Predict whether a user is interested in a target wallpaper.

[0184] If the prediction result is yes, that is, the user is interested in the target wallpaper, then continue to perform the following step S703 to identify the products associated with the wallpaper elements in the target wallpaper;

[0185] If the prediction result is negative, meaning the user is not interested in the target wallpaper, then the process ends.

[0186] It is understood that the prediction process performed in step S702 is an optional process, or optional step. In some embodiments, terminal devices such as mobile phones 10 may also skip the prediction process in step S702 and directly execute the following steps S703 to S704 to display the identification of the associated goods, which is not limited here.

[0187] For example, mobile phone 10 can predict whether a user is interested in a target wallpaper by detecting the user's actions on the displayed target wallpaper. User actions that indicate user interest in the displayed target wallpaper may include actions such as clicking to favorite, taking a screenshot, or downloading or saving the target wallpaper image on a relevant interface (e.g., the lock screen), as well as user actions such as staying on the relevant interface for more than a preset time threshold, etc., without limitation.

[0188] Referring to Figure 6a above, on the wallpaper preview interface 610, the mobile phone 10 can, for example, detect an operation where the user clicks on the more controls 615 and selects the "Save" option, thus saving the target wallpaper image displayed on the wallpaper preview interface 610. Based on the detected save operation, the mobile phone 10 can predict that the user may be interested in the target wallpaper image.

[0189] It is understandable that a user's interest in a target wallpaper may include: the user's interest in the style, color, or combination of products corresponding to the wallpaper elements such as clothing, makeup, and wearable devices in the target wallpaper, as well as the user's interest in the tourist attractions corresponding to the background elements in the target wallpaper, etc., without any restrictions.

[0190] It is understandable that the user operations detected by the phone on the interface displaying the target wallpaper may include operations that indicate that the user is not interested in the target wallpaper. For example, the user may directly enter the unlock password or verify biometric information such as fingerprint on the lock screen to unlock, or click the application icon on the desktop displayed by the phone to use the application, or the user may swipe left or right on the lock screen to switch to displaying another wallpaper. There are no restrictions on these operations.

[0191] S703: Identify the products corresponding to each wallpaper element in the target wallpaper.

[0192] For example, when the mobile phone 10 detects an operation indicating that the user is interested in the target wallpaper during the execution of the above step S702, it can identify the products corresponding to each wallpaper element in the target wallpaper. For example, clothing elements correspond to hats, clothes, shoes, necklaces, bracelets, earrings, etc.; makeup elements correspond to eyebrow pencils, eyeshadows, lipsticks, etc.; wearable device elements correspond to watches, headphones, glasses, etc.; and background elements correspond to tourist attractions and travel guides, etc. There are no restrictions on these.

[0193] It is understandable that the process of recognizing each wallpaper element in the target wallpaper by mobile phone 10 can utilize image segmentation technology to segment each wallpaper element in the target wallpaper image, and based on the image-based search function of the shopping application installed on mobile phone 10, identify the product corresponding to each wallpaper element. Then, mobile phone 10 can continue to perform the following step S704 on each wallpaper element of the target wallpaper, adding relevant product identifiers for the user to click and view related product links.

[0194] It's understandable that the phone can identify the corresponding products for each wallpaper element in a target wallpaper and make intelligent recommendations based on the user's relevant historical behavior data. For example, if multiple shopping apps are installed on the phone, when searching for related products for each wallpaper element based on the image segmentation technology mentioned above, the phone can prioritize recommending product links from frequently used shopping apps based on the user's frequency of use. For another example, if the phone identifies clothing elements as elements from clothing matching images viewed and saved by the user from app A, the phone can prioritize recommending product links for related products provided by app A. Similarly, if the phone identifies background elements as elements from tourist attraction images in travel guides saved by the user in app B, the phone can prioritize recommending related travel services offered by app B, and so on.

[0195] It is understandable that the wallpaper elements identified by the phone 10 for the target wallpaper can be some of the wallpaper elements contained in the target wallpaper, that is, some wallpaper elements that cannot be searched for related products, such as background elements of tourist attractions that cannot be matched. The phone 10 may not search for products corresponding to these elements or add product labels.

[0196] S704: Display product logo on the corresponding wallpaper element.

[0197] For example, after recognizing the product corresponding to the wallpaper element of the target wallpaper, the mobile phone 10 can add a product identifier to the corresponding wallpaper element and display it on the relevant display interface of the target wallpaper, such as on the lock screen or desktop. That is, the mobile phone 10 can display the relevant product identifier on the corresponding wallpaper element while displaying the target wallpaper.

[0198] It is understood that this product icon can be associated with a corresponding product link. When a user clicks on this product icon, the mobile phone 10 can display the purchase interface for the corresponding product. The specific process can be found in the relevant description below, and will not be elaborated upon here.

[0199] S705: Detects whether the user clicks on the product icon.

[0200] If the detection result is yes, that is, the user clicked on the product icon, then the mobile phone 10 can continue to execute the following step S706.

[0201] If the detection result is negative, meaning no user click on the product icon was detected, then the mobile phone 10 can end this process.

[0202] For example, during the process of displaying the target wallpaper, based on the execution results of steps S701 to S704 above, after displaying the product logo on the corresponding wallpaper element, the mobile phone 10 may receive a click operation from the user on the product logo using a touch object, such as a finger. In this case, the mobile phone 10 detects that the user has clicked the product logo, and the corresponding detection result is "yes". Conversely, if the mobile phone 10 does not receive a click operation from the user on the product logo, it does not detect any user operation on the product logo, and the corresponding detection result is "no".

[0203] It is understandable that mobile devices such as the 10 can also obtain user profile data or user preference data as a reference factor for recommending products corresponding to each wallpaper element. This user profile data or user preference data can be determined based on the analysis of user behavior on various applications, and user behavior on various applications can include behavior on one or more terminal devices, without limitation. Furthermore, the detection result of whether the user clicked on the product icon on the 10 can be determined based on the listening results of click events on the product icon on the relevant interface displaying the target wallpaper on the 10's system, which will not be elaborated upon here.

[0204] S706: Display the product purchase interface based on the product link associated with the product identifier.

[0205] For example, mobile phone 10 can open the product purchase interface corresponding to the product link associated with the product icon clicked by the detected user operation. This product purchase interface can be displayed as a floating window, a full-screen window covering the display interface corresponding to the target wallpaper, or a half-screen window covering a portion of the display interface corresponding to the target wallpaper; there are no limitations on this. The half-screen window can, for example, include a window extending upwards from the bottom of the screen of the mobile phone 10 or other terminal device, covering 50% to 90% of the screen display area.

[0206] It is understood that the target wallpaper mentioned above can be a personalized wallpaper generated based on the wallpaper display method provided in this application, a wallpaper randomly recommended by the system of a mobile phone or other terminal device, or an image that the user saves and sets as wallpaper, etc., and there are no restrictions here.

[0207] It is understood that, based on the implementation process of steps S701 to S706 above, the wallpaper display method provided by this application can not only provide users with personalized wallpaper customization functions, but also recommend their favorite products and provide users with relevant product purchase links, which can provide users with more convenient services and improve the user experience.

[0208] Based on the implementation flow of the wallpaper display method shown in Figures 4b and 7 above, it can be understood that the wallpaper display method provided in this application can generate and display personalized wallpapers (or personalized target wallpapers) that meet the user's personalized customization needs, according to the wallpaper material image (e.g., a personal portrait) specified by the user. Furthermore, during the display of personalized wallpapers, mobile phones and other terminal devices can recommend desired products to users based on the wallpaper display method provided in this application, and provide users with shopping services for related products, providing convenience and improving the user experience. In this embodiment, recommending desired products to users can be products identified by the mobile phone or other terminal device from the predicted target wallpapers that the user is interested in.

[0209] Thus, the complete implementation process of the wallpaper display method provided in this application can be seen in Figure 8, including the process of generating "personalized wallpaper" based on "personal portrait photo", the process of recommending products corresponding to the wallpaper elements and displaying product icons when the user is detected to be interested in the displayed personalized wallpaper, and the process of opening the corresponding associated product link to provide "shopping services" to the user when the user clicks on the product icon. Based on the wallpaper display method provided in this application, while providing users with more convenient shopping services, it can also give monetization value to the personalized wallpapers generated based on the wallpaper display method provided in this application.

[0210] The following section, in conjunction with Embodiment 2, describes another specific implementation process of the wallpaper display method provided in this application in a multi-device interaction scenario.

[0211] Example 2

[0212] Referring to the multi-device interaction scenario provided in Figure 4a of Embodiment 1 above, this application embodiment continues to use mobile phone 10 and server 20 as examples to introduce the specific implementation process of a wallpaper display method that generates and displays personalized wallpapers based on user-specified wallpaper elements and wallpaper styles.

[0213] Figure 9 illustrates a schematic diagram of the implementation process of a wallpaper display method according to an embodiment of this application.

[0214] It is understandable that this implementation process may continue to involve the interaction between mobile phone 10 and server 20. Server 20 can be a wallpaper server or a theme server, and it is pre-deployed with a personalized wallpaper generation model for generating personalized wallpapers. This model may include a control network or algorithm capable of extracting corresponding dimensional control information from the wallpaper element images specified by the user, and a diffusion model 330 capable of loading corresponding style parameters according to the wallpaper style specified by the user. Thus, server 20 can configure the diffusion model according to the style parameters corresponding to the wallpaper style specified by the user, and add multi-dimensional wallpaper elements to the target character in the wallpaper material image using the wallpaper elements specified by the user, ultimately generating a target wallpaper image that meets the user's personalized customization requirements for display on mobile phone 10 for the user's selection.

[0215] Specifically, as shown in Figure 9, the implementation process may include the following steps:

[0216] S901: Phone 10 detected the user's action of setting the wallpaper and displayed the relevant settings interface.

[0217] S902: Mobile phone 10 obtains the wallpaper image specified by the user in the relevant settings interface.

[0218] S903: Mobile phone 10 sends wallpaper material images to server 20.

[0219] S904: Server 20 performs portrait analysis on each target person in the wallpaper material image and extracts facial features as facial control information.

[0220] The execution content of steps S901 to S904 above is the same as that of steps S401 to S404 in the above embodiment 1. The specific execution process can be referred to the relevant description of the corresponding steps in the above embodiment 1, and will not be repeated here.

[0221] In this embodiment, after the mobile phone 10 performs step S902 to obtain the wallpaper material image specified by the user, for example, after detecting that the user clicks the "Personalize" control 561 in the wallpaper preview interface shown in Figure 5f, the mobile phone 10 can also display a settings interface that supports user customization of personalized wallpapers. This settings interface allows the user to add other wallpaper element images and select wallpaper styles.

[0222] Referring to Figure 10, taking wallpaper elements such as clothing, pose, and background as examples, the settings interface X10 displayed by the mobile phone 10 can include an add control X11 for adding clothing element images, an add control X12 for adding pose element images, an add control X13 for adding background element images, and more controls X14. Correspondingly, when the mobile phone 10 detects that the user clicks the add control X11, it can display add options such as "Select from gallery" X111 and "Select from camera" X112. "Select from gallery" X111 allows the user to select a saved or photographed image from the gallery to provide clothing elements, while "Select from camera" X112 allows the user to take a photo of clothing of interest in real time to provide clothing elements. Similarly, when the mobile phone 10 detects that the user clicks the add control X12 or add control X13, it can also display options such as "Select from gallery" and "Select from camera" under the corresponding control. The specific implementation process of the mobile phone 10 receiving images related to clothing, pose, and background elements added by the user and providing them to the server 20 for extracting control information of the corresponding dimensions can be found in steps S905 to S913 below. It is understandable that the aforementioned control information in each dimension can be used in the process of generating personalized wallpapers to control the addition of wallpaper elements of corresponding dimensions to the foreground elements such as the target person provided by the wallpaper material image.

[0223] S905: Mobile phone 10 receives images of clothing elements specified by the user.

[0224] For example, the mobile phone 10 can receive the clothing element image specified by the user when it detects the user's relevant operation on the settings interface X10 shown in Figure 10, such as clicking the add control X11 and selecting the specified image to be added.

[0225] S906: Mobile phone 10 sends clothing element images to server 20.

[0226] For example, mobile phone 10 can send the received clothing element image to server 20 so that server 20 can perform the following step S907 to extract clothing control information of the clothing dimension.

[0227] It is understood that the process of receiving the user-specified clothing element image by mobile phone 10 as described in step S905 above, and the process of sending the clothing element image to server 20 as described in step S906 above, are both optional processes, i.e., optional steps. In some other embodiments, mobile phone 10 may not implement the above steps S905 to S906. Correspondingly, server 20 may perform step S907 below to extract clothing control information based on the saved clothing element image, without limitation. The clothing element image saved by server 20 may be, for example, an image containing elements such as clothing of a person uploaded by a user of a shopping application, media application, or other terminal device with such application installed.

[0228] S907: Server 20 extracts clothing control information based on clothing element images.

[0229] Specifically, the process by which server 20 extracts clothing control information based on user-specified clothing element images will be described in detail below with reference to the accompanying drawings, and will not be repeated here.

[0230] S908: Mobile phone 10 receives a user-specified pose element image.

[0231] For example, the mobile phone 10 can receive the image of the gesture element specified by the user when it detects the user's operation on the settings interface X10 shown in Figure 10, such as clicking the add control X12 and selecting the image to be added.

[0232] S909: Mobile phone 10 sends attitude element images to server 20.

[0233] For example, the mobile phone 10 can send the received pose element image to the server 20 so that the server 20 can perform the following step S910 to extract the pose control information of the pose dimension.

[0234] It is understood that the process of receiving the user-specified posture element image by mobile phone 10 as described in step S908 above, and the process of sending the posture element image to server 20 as described in step S909 above, are both optional processes, i.e., optional steps. In some other embodiments, mobile phone 10 may not implement the above steps S908 to S909. Correspondingly, server 20 may perform step S910 below to extract posture control information based on the saved posture element image, without limitation. The posture element image saved by server 20 may be, for example, an image containing elements such as the posture of a person uploaded by a user of a shopping application, media application, or other terminal device with such application installed.

[0235] S910: Server 20 extracts attitude control information based on attitude element images.

[0236] Specifically, the process by which server 20 extracts attitude control information based on the user-specified attitude element image will be described in detail below with reference to the accompanying drawings, and will not be repeated here.

[0237] S911: Mobile phone 10 receives a background element image specified by the user.

[0238] For example, the mobile phone 10 can receive the background element image specified by the user when it detects the user's relevant operation on the settings interface X10 shown in Figure 10, such as clicking the add control X13 and selecting the specified image to be added.

[0239] S912: Mobile phone 10 sends background element images to server 20.

[0240] For example, mobile phone 10 can send the received background element image to server 20 so that server 20 can perform the following step S913 to extract background control information of the background dimension.

[0241] It is understood that the process of receiving the user-specified background element image by mobile phone 10 as described in step S911 above, and the process of sending the background element image to server 20 by mobile phone 10 as described in step S912 above, are both optional processes, i.e., optional steps. In some other embodiments, mobile phone 10 may not implement the above steps S911 to S912. Correspondingly, server 20 may perform step S913 below to extract background control information based on the saved background element image, without limitation. The background element image saved by server 20 may be, for example, an image containing scenic spots or buildings as backgrounds uploaded by users of shopping applications, media applications, etc., through terminal devices with such applications installed.

[0242] S913: Server 20 obtains background control information based on background element images.

[0243] Specifically, the process by which server 20 extracts background control information based on the background element image specified by the user will be described in detail below with reference to the relevant accompanying drawings, and will not be repeated here.

[0244] It is understood that in other embodiments, the mobile phone 10 may also receive wallpaper element images of other dimensions specified by the user and send them to the server 20 so that the server 20 can extract control information of the corresponding dimensions, thereby adding multi-dimensional wallpaper elements to foreground elements such as target figures in the wallpaper material image. This is not limited here.

[0245] S914: Mobile phone 10 receives wallpaper styles specified by the user.

[0246] For example, the settings interface displayed on mobile phone 10, which supports user-customized personalized wallpapers, can receive user-specified wallpaper styles. As an example, the settings interface X10 shown in Figure 10 can also include controls for users to specify wallpaper styles, such as a selection control X15 corresponding to "Please select a wallpaper style," and various style controls displayed when the user clicks on the selection control X15, including but not limited to "futuristic style," "dreamy style," "ancient style," "retro style," "modern style," "fashion magazine style," "academic style," "pastoral style," "oil painting style," etc. For example, mobile phone 10 can receive the user-specified "pastoral style" wallpaper style in response to the user clicking on the style control X151 corresponding to "pastoral style." Correspondingly, based on the user-specified "pastoral style" wallpaper style, mobile phone 10 can determine the information identifier corresponding to that wallpaper style.

[0247] It is understood that the mobile phone 10 can pre-store information identifiers corresponding to various wallpaper styles specified by the user, so as to match the corresponding information identifiers with the received user-specified wallpaper styles. After the information identifiers are provided to the server 20, the server 20 can load the corresponding style parameter configuration diffusion model 330, or generate the corresponding style prompt information input to the diffusion model 330. See the description of the relevant steps below for details, which will not be repeated here.

[0248] S915: Mobile phone 10 sends the information identifier corresponding to the wallpaper style specified by the user to server 20.

[0249] For example, after determining the corresponding information identifier based on the wallpaper style specified by the user, the mobile phone 10 can send the information identifier to the server 20.

[0250] It is understood that the process of receiving the user-specified wallpaper style described in step S914 above, and the process of sending the wallpaper style to the server 20 described in step S915 above, are both optional processes, i.e., optional steps. In some other embodiments, the mobile phone 10 may not implement the above steps S914 to S915. Correspondingly, the server 20 may execute the following step S916 to load the style control information corresponding to the wallpaper style based on the information identifier corresponding to the wallpaper style matched by matching the user behavior data of the mobile phone 10 or by model inference. This is not limited here.

[0251] S916: Server 20 loads style control information indicated by information identifiers, configuring or inputting diffusion models.

[0252] For example, server 20 can load corresponding style parameters based on the deployed personalized wallpaper generation model and the received information identifier indicating the wallpaper style, and then obtain the style parameters to configure diffusion model 330. Server 20 can also load a pre-trained prompt generation model based on the received information identifier to generate prompt information corresponding to the style indicated by the information identifier, and input it into diffusion model 330.

[0253] It is understood that server 20 can preset multiple trained and determined style parameters, each corresponding to a different wallpaper style. In some embodiments, server 20 can be configured with a model library for storing each style parameter, hereinafter referred to as a stylized model library, for use by the personalized wallpaper generation model to load the style parameters of the corresponding wallpaper style. Server 20 can also preset a prompt generation model capable of generating corresponding prompt information based on multiple different styles; this is not limited here.

[0254] S917: Server 20 inputs face control information, clothing control information, posture control information, and background control information into the configured diffusion model to generate a personalized target wallpaper image.

[0255] S918: Server 20 sends the target wallpaper image to mobile phone 10.

[0256] S919: Phone 10 displays the target wallpaper preview interface.

[0257] S920: The phone responds to a user's action of setting a target wallpaper image as the wallpaper and completes the wallpaper setting.

[0258] S921: Mobile phone 10 displays the target wallpaper in response to user actions that instruct it to display the lock screen or desktop interface.

[0259] The execution content of steps S917 to S918 above is the same as the execution content of steps S409 to S413 in the above embodiment 1. The specific execution process can be referred to the relevant description of the corresponding steps in the above embodiment 1, and will not be repeated here.

[0260] Based on the interactive implementation process of steps S901 to S921 above, the wallpaper display method provided by this application can provide users of terminal devices such as mobile phones 10 with the functions of customizing and displaying personalized wallpapers. For example, according to the wallpaper material image, wallpaper element image and wallpaper style specified by the user, the server 20 is called to add multiple dimensions of wallpaper elements specified by the user to the foreground elements (such as one or more target figures) in the wallpaper material image, and finally display a personalized wallpaper with diversified wallpaper elements, which can be selected by the user to set as the lock screen interface or desktop wallpaper, etc., thus meeting the user's needs for customization or use of personalized wallpapers.

[0261] Correspondingly, based on the interactive implementation process of steps S901 to S921 described above, the terminal device such as mobile phone 10 can also display product icons corresponding to various wallpaper elements on the interface that ultimately displays the personalized target wallpaper image, based on the prediction of whether the user is interested in the wallpaper, and can recommend purchase links for desired products to the user. For a detailed description of the implementation process of the wallpaper display method with product recommendations on the terminal device such as mobile phone 10, please refer to the relevant description of the execution steps in the implementation flow shown in Figure 7 of Embodiment 1 above, which will not be repeated here.

[0262] The following section, in conjunction with Embodiment 3, describes the specific implementation process of the wallpaper display method provided in this application in a single-device implementation scenario.

[0263] Example 3

[0264] In this embodiment, taking mobile phone 10 as an example, the personalized wallpaper generation model for user-generated personalized wallpapers can be pre-deployed on terminal devices such as mobile phone 10. Correspondingly, terminal devices such as mobile phone 10 can have strong processing capabilities to support the operation of the personalized wallpaper generation model. Thus, in a single-device implementation scenario that only includes terminal devices such as mobile phone 10, the wallpaper display method provided in this application can also be implemented on the terminal devices such as mobile phone 10.

[0265] As mentioned earlier, the model may include a control network or algorithm capable of extracting control information of corresponding dimensions based on default recommended or user-specified wallpaper element images, and a diffusion model 330 capable of loading style parameters based on default recommended or user-specified wallpaper styles. Thus, the mobile phone 10 can configure the diffusion model 330 according to the user-specified wallpaper style, and use default recommended or user-specified wallpaper elements to add or adjust multiple dimensions of wallpaper elements for the target person in the wallpaper material image, ultimately generating a target wallpaper image that meets the user's personalized customization requirements for display on the mobile phone 10.

[0266] Figure 11 illustrates a schematic diagram of the implementation process of a wallpaper display method according to an embodiment of this application.

[0267] It is understood that in the embodiments of this application, the execution subject of each step in the implementation process shown in FIG11 can be mobile phone 10. In other embodiments, the execution subject of each step in the implementation process shown in FIG11 can also be other terminal devices with strong processing capabilities, such as tablet computers, laptops, etc., which are not limited here.

[0268] Specifically, as shown in Figure 11, the implementation process may include the following steps:

[0269] S101: A user action to set the wallpaper was detected, and the relevant settings interface was displayed.

[0270] S102: Obtain the wallpaper image specified by the user in the relevant settings interface.

[0271] The process of mobile phone 10 performing the above steps S101 to S102 can be referred to the relevant description of the execution process of steps S401 to S402 in the above embodiment 1, which will not be repeated here.

[0272] S103: Perform portrait analysis on each target person in the wallpaper material image and extract facial features as facial control information.

[0273] The process of mobile phone 10 executing step S103 can be referred to in the relevant description of server 20 executing step S403 in the above embodiment 1. Based on the portrait analysis module 310 integrated in the pre-deployed personalized wallpaper generation model, the facial features corresponding to each person in the wallpaper material image are extracted, and the facial control information of the target person is generated according to the extracted facial feature information.

[0274] S104: Obtain clothing control information based on default recommended or user-specified clothing element images.

[0275] For example, the mobile phone 10 can determine the clothing control information used to control the personalized wallpaper generation result based on the feature vector or feature vector set of the default recommended clothing elements, or it can extract the corresponding clothing control information based on the clothing element image specified by the user on the relevant settings interface.

[0276] The process of obtaining clothing control information by mobile phone 10 according to the default recommendation can be referred to the relevant description of the execution step S405 of server 20 in the above embodiment 1, and will not be repeated here.

[0277] The process of mobile phone 10 obtaining clothing control information based on the clothing element image specified by the user can be referred to the relevant description of server 20 executing step S907 in the above embodiment 2. The process of mobile phone 10 obtaining the clothing element image specified by the user can be referred to the relevant description of mobile phone 10 executing step S905 in the above embodiment 2, and will not be repeated here.

[0278] S105: Obtain attitude control information based on the default recommended or user-specified attitude element images.

[0279] For example, the mobile phone 10 can determine the posture control information used to control the personalized wallpaper generation result based on the feature vector or feature vector set of the default recommended posture map, or it can extract the corresponding posture control information based on the posture element image specified by the user on the relevant settings interface.

[0280] The process of obtaining attitude control information by mobile phone 10 according to the default recommendation can be referred to the relevant description of the execution step S406 of server 20 in the above embodiment 1, and will not be repeated here.

[0281] The process of mobile phone 10 obtaining attitude control information based on the attitude element image specified by the user can be referred to the relevant description of server 20 executing step S910 in the above embodiment 2. The process of mobile phone 10 obtaining the attitude element image specified by the user can be referred to the relevant description of mobile phone 10 executing step S908 in the above embodiment 2, and will not be repeated here.

[0282] S106: Obtain background control information based on the default recommended or user-specified background element image.

[0283] For example, the mobile phone 10 can determine the background control information used to control the personalized wallpaper generation result based on the feature vector or feature vector set corresponding to the default recommended background element, or it can extract the corresponding background control information based on the background element image specified by the user on the relevant settings interface.

[0284] The process of obtaining background control information by mobile phone 10 according to the default recommendation can be referred to the relevant description of the execution step S407 of server 20 in the above embodiment 1, and will not be repeated here.

[0285] The process of mobile phone 10 obtaining background control information based on the background element image specified by the user can be referred to the relevant description of the server 20 executing step S913 in the above embodiment 2. The process of mobile phone 10 obtaining the background element image specified by the user can be referred to the relevant description of the mobile phone 10 executing step S911 in the above embodiment 2, and will not be repeated here.

[0286] S107: Determine the information identifier for the corresponding style based on the default recommendation or the user-specified wallpaper style.

[0287] It is understood that the processes described in steps S104 to S106 above for obtaining relevant control information based on default recommendations or user-specified elements such as clothing, posture, and background, and the process described in step S107 above for determining the information identifier of the corresponding style based on the default recommendations or user-specified wallpaper style, are all optional steps. That is, one or more of the above steps S104 to S106 and S107 can be implemented. In some other embodiments, the mobile phone 10 may not implement the above steps S104 to S106 and S107, or may only implement one or more of them, and combine the collected user behavior data to match or match the wallpaper style through model inference, and match elements such as clothing, posture, and background. Then, step S108 is executed to input the corresponding style control information into the diffusion model, and step S109 is executed to input the corresponding face control information, clothing control information, posture control information, and background control information into the configured diffusion model to generate a personalized wallpaper. No restrictions are imposed here.

[0288] S108: Load style control information indicated by information identifiers, configure or input diffusion models.

[0289] For example, the mobile phone 10 can load the corresponding style parameters and configure the diffusion model 330 based on the default recommended wallpaper style. Alternatively, the mobile phone 10 can also determine the corresponding information identifier based on the wallpaper style selected by the user in the relevant settings interface, and then load the prompt generation model to generate the prompt information corresponding to the style indicated by the identifier, and input it into the diffusion model 330.

[0290] The execution process of loading the default recommended style parameter configuration diffusion model on mobile phone 10, or generating corresponding style prompt information input into the diffusion model, can be referred to the relevant description of the server 20 execution step S408 in the above embodiment 1, and will not be repeated here.

[0291] The execution process of determining the information identifier corresponding to the style control information based on the wallpaper style selected by the user on the relevant settings interface can be referred to the relevant description of step S914 executed by mobile phone 10 in embodiment 2 above. The execution process of mobile phone 10 loading the style parameter configuration diffusion model indicated by the information identifier or generating the corresponding style prompt information input diffusion model can be referred to the relevant description of step S916 executed by server 20 in embodiment 2 above above, and will not be repeated here.

[0292] S109: Input the face control information, clothing control information, posture control information, and background control information into the configured diffusion model to generate a personalized target wallpaper image.

[0293] The process of mobile phone 10 executing this step S109 can be referred to the relevant description of the execution process of server 20 executing step S409 in the above embodiment 1, and will not be repeated here.

[0294] S110: Displays the target wallpaper preview interface.

[0295] S111: In response to a user action to set the target wallpaper image as the wallpaper, complete the wallpaper setting.

[0296] S112: In response to a user action that instructs the display of an interface such as the lock screen or desktop, display the target wallpaper.

[0297] The execution content of steps S110 to S112 above is the same as the execution content of steps S409 to S413 in the above embodiment 1. The specific execution process can be referred to the relevant description of the corresponding steps in the above embodiment 1, and will not be repeated here.

[0298] Based on the interactive implementation process of steps S901 to S921 above, the wallpaper display method provided by this application can provide users of terminal devices such as mobile phones 10 with the functions of customizing and displaying personalized wallpapers. For example, according to the wallpaper material image, wallpaper element image and wallpaper style specified by the user, multiple dimensions of wallpaper elements specified by the user are added to the foreground elements (such as the faces of one or more target people) in the wallpaper material image, and finally a personalized wallpaper with diversified wallpaper elements is displayed for the user to choose to set as the lock screen interface or desktop wallpaper, etc., thus meeting the user's needs for customization or use of personalized wallpapers.

[0299] Correspondingly, based on the implementation process of steps S101 to S112 described above, the terminal device such as mobile phone 10 can also display product icons corresponding to various wallpaper elements on the interface that ultimately displays the personalized target wallpaper image, based on the prediction of whether the user is interested in the wallpaper, and can recommend purchase links for desired products to the user. For a detailed description of the implementation process of the wallpaper display method with product recommendations by the terminal device such as mobile phone 10, please refer to the relevant description of the execution steps in the implementation process shown in Figure 7 of Embodiment 1 above, which will not be repeated here.

[0300] The following describes in detail the specific implementation process of determining the control information of dimensions such as face, clothing, posture, and background involved in Embodiments 1 to 3 above, with reference to the relevant accompanying drawings.

[0301] Figure 12a illustrates a schematic diagram of the process for determining clothing control information according to an embodiment of this application.

[0302] As shown in Figure 12a, the process of determining clothing control information may include steps a201 to a204:

[0303] a201: Default Recommendation. This refers to the clothing control module 322 integrated into the personalized wallpaper generation model, which can obtain default recommended clothing element images.

[0304] It is understandable that mobile phones 10 and other terminal devices or servers 20 can preset and support default recommendations of diverse and fashionable clothing element images. These clothing element images can include clothing elements such as hats, clothes, shoes, necklaces, bracelets, and earrings, and can present the matching effect when these clothing elements are combined.

[0305] a202: User-specified. This refers to the clothing control module 322 integrated into the personalized wallpaper generation model, which can acquire user-specified clothing element images. These user-specified clothing element images can include images uploaded by the user from which clothing features can be extracted, or images specified by the user from multiple default recommended clothing element images displayed on the relevant settings interface of a mobile device such as the 10th generation.

[0306] a203: User selection. This refers to the clothing control module 322 integrated into the personalized wallpaper generation model, which can determine the user's selected method for acquiring the clothing element image.

[0307] It is understandable that if the user selects to upload or specify an image of clothing elements on the relevant settings interface, the clothing control module 322 can determine that the user has selected the "user-specified" acquisition method provided in step a202 above. Conversely, if the user does not upload or specify an image of clothing elements on the relevant settings interface, the clothing control module 322 can determine that the user has selected the "default recommendation" acquisition method provided in step a201 above.

[0308] a204: Input control network. This refers to the clothing control module 322 integrated into the personalized wallpaper generation model. It can input clothing element images obtained through a user-selected acquisition method into a pre-trained control network, causing the control network to output clothing control information.

[0309] It is understood that the pre-trained control network described above can be a neural network model capable of outputting clothing control information based on the input clothing element image. For example, this control network can extract feature vectors or sets of feature vectors including one or more clothing elements from the image data of the clothing element image and output them as clothing control information.

[0310] It is understood that, based on the specific implementation process described in steps a201 to a204 of Figure 12a above, in the implementation of the wallpaper display method provided in this application, it is possible to obtain clothing control information of wallpaper elements for controlling the clothing dimension for the face of the target person in the wallpaper material image specified by the user.

[0311] It is understood that the personalized generation model deployed on terminal devices such as mobile phones 10 or servers 20 in the above embodiments 1 to 3 can integrate clothing control modules 322 that can implement the above steps a201 to a204.

[0312] Figure 12b illustrates a schematic diagram of the process for determining attitude control information according to an embodiment of this application.

[0313] As shown in Figure 12b, the process of determining attitude control information may include steps b201 to b205:

[0314] b201: Pose estimation. This refers to the pose control module 323 integrated into the personalized wallpaper generation model, which can estimate the pose of a person in a pose element image uploaded by the user. In this embodiment, the pose estimation process may include first identifying the relative positions of key points of the human skeleton of the target person in the pose element image, and then drawing a pose diagram of the target person based on the relative positions of these key points.

[0315] b202: Attitude map. This refers to the attitude control module 323 integrated into the personalized wallpaper generation model, which can draw the attitude map corresponding to the user-specified attitude element image based on the attitude estimation results mentioned above.

[0316] b203: Pose Template Library. This refers to the pose control module 323 integrated into the personalized wallpaper generation model, which can retrieve pose images used to control the pose of the target character from a preset pose template library. This retrieval process can be based on default recommendations, including random or sequential recommendations, etc., and is not limited here.

[0317] Referring to Figure 12b, the preset pose images in the pose template library can correspond to various character poses. Furthermore, the pose template library can be updated based on the pose images obtained in steps b201 to b202 above. For example, the pose images obtained in steps b201 to b202 above can be added to the pose template library as one of the recommended pose images.

[0318] b204: User selection. This refers to the posture control module 323 integrated into the personalized wallpaper generation model, which can determine the user's selection of the method for obtaining the posture map.

[0319] It is understandable that if the user selects to upload or specify a pose element image on the relevant settings interface, the pose control module 323 can determine that the user has selected the "user-specified" pose element image obtained through steps b201 to b202 above, and the pose map is determined through pose estimation. Conversely, if the user does not upload or specify a pose element image on the relevant settings interface, the clothing control module 322 can determine that the user has selected the default method from the "pose template library" provided in step b203 above.

[0320] b205: Input control network. This refers to the posture control module 323 integrated into the personalized wallpaper generation model. It can input posture element images obtained through a user-selected acquisition method into a pre-trained control network, causing the control network to output posture control information.

[0321] It is understood that the pre-trained control network described above can be a neural network model capable of outputting posture control information based on the input posture map. For example, this control network can extract feature vectors or sets of feature vectors of key points of the human skeleton from the image data of the posture map and output them as posture control information.

[0322] It is understood that, based on the specific implementation process described in steps b201 to b205 of the example in Figure 12b above, in the implementation of the wallpaper display method provided in this application, it is possible to obtain posture control information for adjusting the posture of the target person in the wallpaper material image specified by the user.

[0323] Figure 12c illustrates a schematic diagram of the process for determining background control information according to an embodiment of this application.

[0324] As shown in Figure 12c, the process of determining the background control information may include steps c201 to c204:

[0325] c201: Default Recommendation. This refers to the background control module 324 integrated into the personalized wallpaper generation model, which can obtain default recommended background element images.

[0326] It is understandable that mobile phones 10 and other terminal devices or servers 20 can preset and support a variety of fashionable background element images that are recommended by default. These background element images may include background elements such as flowers, mountains and rivers, trees and forests, or buildings, historical sites, etc.

[0327] c202: User-specified. This refers to the background control module 324 integrated into the personalized wallpaper generation model, which can acquire user-specified background element images. These user-specified background element images can include user-uploaded background element images with extractable background features, or background element images specified by the user from multiple default recommended background element images displayed on the relevant settings interface of a mobile device such as the 10th generation.

[0328] c203: User selection. This refers to the background control module 324 integrated into the personalized wallpaper generation model, which can determine the method of obtaining the background element image selected by the user.

[0329] It is understandable that if the user selects to upload or specify a background element image on the relevant settings interface, the background control module 324 can determine that the user has selected the "user-specified" acquisition method provided in step c202 above. Conversely, if the user does not upload or specify a background element image on the relevant settings interface, the background control module 324 can determine that the user has selected the "default recommended" acquisition method provided in step c201 above.

[0330] c204: Input control network. This refers to the background control module 324 integrated into the personalized wallpaper generation model. It can input background element images obtained through a user-selected acquisition method into a pre-trained control network, causing the control network to output background control information.

[0331] It can be understood that the pre-trained control network described above can be a neural network model capable of outputting background control information based on the input background element image. For example, this control network can extract feature vectors or sets of feature vectors including one or more background elements from the image data of the background element image and output them as background control information.

[0332] It is understood that, based on the specific implementation process described in steps c201 to c204 of the example in Figure 12c above, in the implementation of the wallpaper display method provided in this application, background control information of wallpaper elements for controlling the background dimension can be obtained for the target person in the wallpaper material image specified by the user.

[0333] It is understood that the personalized generation model deployed on terminal devices such as mobile phones 10 or servers 20 in the above embodiments 1 to 3 can integrate a background control module 324 capable of implementing the above steps c201 to c204.

[0334] The following section, in conjunction with the accompanying drawings, details the specific implementation process of configuring the diffusion model using the loading style parameters involved in Embodiments 1 to 3.

[0335] Figure 12d illustrates a schematic diagram of the implementation process of configuring a diffusion model with loading style parameters according to an embodiment of this application.

[0336] As shown in Figure 12d, the process of configuring the diffusion model by loading style parameters can include steps d201 to d205:

[0337] d201: Stylization Model Library. This refers to the diffusion model 330 integrated into the personalized wallpaper generation model. It can load corresponding style parameters from a pre-trained stylization model library and configure the diffusion model 330. The style parameters stored in this stylization model library can be pre-trained parameter templates that can configure the diffusion model 330 to generate personalized wallpapers of corresponding styles.

[0338] It is understandable that the style parameters stored in the aforementioned stylization model library can correspond to styles including futuristic, dreamlike, ancient, retro, modern, fashion magazine, academic, pastoral, and oil painting styles. The style parameters loaded by diffusion model 330 can be loaded based on the default recommended method (see step d202 below for details), or based on a user-specified method (see step d203 below for details), which will not be elaborated upon here.

[0339] d202: Default Recommendation. This refers to the diffusion model 330 integrated into the personalized wallpaper generation model, which can load the default recommended style parameters based on the information identifiers corresponding to the default recommended style.

[0340] It is understood that the above default recommendation method can include random recommendation or sequential recommendation, etc., and there are no restrictions here.

[0341] d203: User-specified. This refers to the diffusion model 330 integrated into the personalized wallpaper generation model. Based on the wallpaper style selected by the user in the relevant settings interface, it determines the corresponding information identifier and then loads the default recommended style parameters. In some embodiments, the diffusion model 330 can also convert and generate corresponding prompt information as input based on the wallpaper style selected by the user in the relevant interface; this is not limited here.

[0342] It is understandable that users can choose the wallpaper style settings interface, and can refer to the settings interface shown in Figure 10 above X10 or other styles of settings interfaces, without any restrictions here.

[0343] d204: User selection. This refers to the diffusion model 330 integrated into the personalized wallpaper generation model, which can determine the loading method corresponding to the style parameters selected by the user.

[0344] It is understandable that if the user selects a wallpaper style in the relevant settings interface, the diffusion model 330 can determine that the user has selected the "user-specified" loading method provided in step d203 above. Conversely, if the user does not select a wallpaper style in the relevant settings interface, the clothing control module 322 can determine that the user has selected the "default recommended" loading method provided in step d202 above.

[0345] d205: Configuration Model. This refers to the diffusion model 330 integrated into the personalized wallpaper generation model. The relevant parameters in the diffusion model 330 can be configured using the loaded style parameters.

[0346] It can be understood that the relevant parameters in the model configured by the style parameters in the diffusion model 330 can be the relevant parameters in the diffusion model 330 used to control the style presented by the final generated personalized wallpaper.

[0347] It is understood that, based on the specific implementation process described in steps d201 to d205 of the example in Figure 12d above, in the implementation of the wallpaper display method provided in this application, it is possible to configure the relevant parameters of the diffusion model 330 that presents the corresponding style for the personalized wallpaper generated based on the wallpaper material image specified by the user.

[0348] It is understood that the personalized generation models deployed on terminal devices such as mobile phones 10 or servers 20 in the above embodiments 1 to 3 can integrate diffusion models 330 that can implement the above steps d201 to d205.

[0349] It is understood that, in various embodiments of this application, terminal devices such as mobile phones 10 that implement the wallpaper display method provided in this application may have the hardware structure exemplified in FIG13 below.

[0350] Figure 13 shows a schematic diagram of the hardware structure of a mobile phone according to an embodiment of this application.

[0351] The mobile phone 10 may include a processor 110, an external memory interface 120, an internal memory 121, a universal serial bus (USB) interface 130, a charging management module 140, a power management module 141, a battery 142, an antenna 1, an antenna 2, a mobile communication module 150, a wireless communication module 160, an audio module 170, a speaker 170A, a receiver 170B, a microphone 170C, a headphone jack 170D, a sensor module 180, buttons 190, a motor 191, an indicator 192, a camera 193, a display screen 194, and a subscriber identity module (SIM) card interface 195, etc. The sensor module 180 may include a pressure sensor 180A, a gyroscope sensor 180B, a barometric pressure sensor 180C, a magnetic sensor 180D, an accelerometer sensor 180E, a distance sensor 180F, a proximity sensor 180G, a fingerprint sensor 180H, a temperature sensor 180J, a touch sensor 180K, an ambient light sensor 180L, a bone conduction sensor 180M, etc.

[0352] It is understood that the structures illustrated in the embodiments of this application do not constitute a specific limitation on the mobile phone 10. In other embodiments of this application, the mobile phone 10 may include more or fewer components than illustrated, or combine some components, or split some components, or have different component arrangements. The illustrated components may be implemented in hardware, software, or a combination of software and hardware.

[0353] Processor 110 may include one or more processing units, such as application processors (APs), modem processors, graphics processing units (GPUs), image signal processors (ISPs), controllers, video codecs, digital signal processors (DSPs), baseband processors, and / or neural network processors (NPUs). These different processing units may be independent devices or integrated into one or more processors.

[0354] The controller can generate operation control signals based on the instruction opcode and timing signals to complete the control of instruction fetching and execution.

[0355] In this embodiment, the processor 10 of the mobile phone 10 can generate operation control signals through the controller to control the reading and execution instructions corresponding to the implementation steps of the wallpaper display method provided in this application, so as to realize the wallpaper display method provided in this application. The implementation steps of the wallpaper display method can refer to the steps shown in Figure 4 of Embodiment 1, Figure 9 of Embodiment 2, and Figure 11 of Embodiment 3, and are not limited here.

[0356] The processor 110 may also include a memory for storing instructions and data. In some embodiments, the processor 110 may include one or more interfaces. Interfaces may include a SIM card interface, and / or a universal serial bus (USB) interface, etc.

[0357] The charging management module 140 receives charging input from the charger. While charging the battery 142, the charging management module 140 can also supply power to the mobile phone 10 through the power management module 141.

[0358] The wireless communication function of mobile phone 10 can be implemented through antenna 1, antenna 2, mobile communication module 150, wireless communication module 160, modem processor, and baseband processor. In some embodiments, antenna 1 of mobile phone 10 is coupled to mobile communication module 150, and antenna 2 is coupled to wireless communication module 160, enabling mobile phone 10 to communicate with networks and other devices through wireless communication technology.

[0359] The mobile phone 10 implements display functions through a GPU, a display screen 194, and an application processor. The GPU is a microprocessor for image processing, connected to the display screen 194 and the application processor. The GPU performs mathematical and geometric calculations for graphics rendering. The processor 110 may include one or more GPUs, which execute program instructions to generate or modify display information. The display screen 194 is used to display images, videos, etc. The display screen 194 includes a display panel. In some embodiments, the mobile phone 10 may include one or N displays 194, where N is a positive integer greater than 1.

[0360] An NPU (Neural Processing Unit) is a neural network (NN) computing processor. By borrowing the structure of biological neural networks, such as the transmission patterns between neurons in the human brain, it can quickly process input information and continuously learn itself. The NPU enables intelligent cognitive applications on the mobile phone 10, such as image recognition, face recognition, speech recognition, and text understanding. In this embodiment, the mobile phone 10, based on the NPU and the aforementioned CPU, can implement the wallpaper display method provided in this application.

[0361] The external memory interface 120 can be used to connect an external memory card, such as a Micro SD card, to expand the storage capacity of the mobile phone 10. The external memory card communicates with the processor 110 through the external memory interface 120 to perform data storage functions, such as storing images.

[0362] Internal memory 121 can be used to store computer executable program code, which includes instructions. Processor 110 executes various functional applications and data processing of mobile phone 10 by running instructions stored in internal memory 121 and / or instructions stored in memory located in the processor.

[0363] The phone 10 can perform audio functions, such as music playback and recording, through an audio module 170, speaker 170A, receiver 170B, microphone 170C, headphone jack 170D, and application processor.

[0364] Buttons 190 include a power button, volume buttons, etc. Buttons 190 can be mechanical buttons or touch buttons. Motor 191 can generate vibration to indicate. Indicator 192 can be an indicator light, used to indicate charging status, battery level changes, messages, missed calls, notifications, etc.

[0365] The SIM card interface 195 is used to connect a SIM card. In some embodiments, the mobile phone 10 uses an eSIM, i.e., an embedded SIM card.

[0366] Figure 14 is a schematic diagram of the structure of a server according to an embodiment of this application.

[0367] As shown in FIG14, in some embodiments, server 20 may include one or more processors 1404, system control logic 1408 connected to at least one of the processors 1404, system memory 1412 connected to system control logic 1408, non-volatile memory (NVM) 1416 connected to system control logic 1408, and network interface 1420 connected to system control logic 1408.

[0368] In some embodiments, processor 1404 may include one or more single-core or multi-core processors. In some embodiments, processor 1404 may include any combination of general-purpose processors and special-purpose processors (e.g., graphics processors, application processors, baseband processors, etc.).

[0369] In some embodiments, system control logic 1408 may include any suitable interface controller to provide any suitable interface to at least one of the processors 1404 and / or any suitable device or component communicating with system control logic 1408.

[0370] In some embodiments, system control logic 1408 may include one or more memory controllers to provide an interface to system memory 1412. System memory 1412 may be used to load and store data and / or instructions. In some embodiments, the memory 1412 of server 20 may include any suitable volatile memory, such as suitable dynamic random access memory (DRAM).

[0371] The non-volatile memory 1416 may include one or more tangible, non-transitory readable storage media for storing data and / or instructions. In some embodiments, the non-volatile memory 1416 may include any suitable non-volatile memory such as flash memory and / or any suitable non-volatile storage device, such as at least one of a hard disk drive (HDD), a compact disc (CD) drive, and a digital versatile disc (DVD) drive.

[0372] The non-volatile memory 1416 may include a portion of the storage resources on the device on which the server 20 is installed, or it may be accessible by the device, but is not necessarily part of the device. For example, the non-volatile memory 1416 may be accessed over a network via network interface 1420.

[0373] System memory 1412 and non-volatile memory 1416 may each include a temporary copy and a permanent copy of instructions 1424. Instructions 1424 may include instructions that, when executed by at least one of processors 1404, cause server 20 to perform the relevant steps in the method shown in FIG. 4b or FIG. 9 above. In some embodiments, instructions 1424, hardware, firmware, and / or their software components may additionally / alternatively be located in system control logic 1408, network interface 1420, and / or processor 1404.

[0374] Network interface 1420 may include a transceiver for providing a radio interface to server 20, thereby enabling communication with any other suitable devices (such as front-end modules, antennas, etc.) over one or more networks. In some embodiments, network interface 1420 may be integrated into other components of server 20.

[0375] The network interface 1420 may further include any suitable hardware and / or firmware to provide a multiple-input multiple-output radio interface. For example, the network interface 1420 may be a network adapter, a wireless network adapter, a telephone modem, and / or a wireless modem.

[0376] In one embodiment, at least one of the processors 1404 may be packaged together with the logic of one or more controllers for system control logic 1408 to form a system initiation protocol (SiP). In another embodiment, at least one of the processors 1404 may be integrated on the same die with the logic of one or more controllers for system control logic 1408 to form a system on chip (SoC).

[0377] Server 20 may further include: input / output (I / O) device 1432. I / O device 1432 may include a user interface that enables a user to interact with server 20; the peripheral component interface is designed so that peripheral components can also interact with server 20.

[0378] This application also provides a computer program product for implementing the wallpaper display methods provided in the above embodiments.

[0379] Various embodiments of the mechanisms disclosed in this application can be implemented in hardware, software, firmware, or combinations of these implementation methods. Embodiments of this application can be implemented as computer program modules or module code executable on a programmable system, the programmable system including at least one processor, a storage system (including volatile and non-volatile memory and / or storage elements), at least one input device, and at least one output device.

[0380] Computer program modules or module code can be applied to input instructions to perform the functions described in this application and generate output information. The output information can be applied to one or more output devices in a known manner. For the purposes of this application, the processing system includes any system having a processor such as, for example, a digital signal processor (DSP), a microcontroller, an application-specific integrated circuit (ASIC), or a microprocessor.

[0381] Module code can be implemented using a high-level modular language or an object-oriented programming language to communicate with the processing system. Assembly language or machine language can also be used to implement module code when needed. In fact, the mechanisms described in this application are not limited to any particular programming language. In either case, the language can be a compiled language or an interpreted language.

[0382] In some cases, the disclosed embodiments may be implemented in hardware, firmware, software, or any combination thereof. The disclosed embodiments may also be implemented as instructions carried or stored on one or more temporary or non-temporary machine-readable (e.g., computer-readable) media, which may be read and executed by one or more processors. For example, the instructions may be distributed via a network or via other computer-readable media. Therefore, machine-readable media may include any mechanism for storing or transmitting information in a machine-readable (e.g., computer-readable) form, including but not limited to floppy disks, optical disks, optical discs, magneto-optical disks, read-only memory (ROM), random access memory (RAM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), magnetic cards or optical cards, flash memory, or tangible machine-readable storage for transmitting information (e.g., carrier waves, infrared signals, digital signals, etc.) using the Internet in the form of electrical, optical, acoustic, or other forms of propagated signals. Therefore, machine-readable media include any type of machine-readable medium suitable for storing or transmitting electronic instructions or information in a machine-readable (e.g., computer-readable) form.

[0383] In this specification, the reference to "an embodiment" or "an embodiment" means that a specific feature, structure, or characteristic described in connection with the embodiment is included in at least one exemplary implementation or technology disclosed according to an embodiment of this application. The appearance of the phrase "in an embodiment" in various places in the specification does not necessarily refer to the same embodiment.

[0384] The disclosure of embodiments of this application also relates to means for performing the operations described herein. Such means may be constructed specifically for the claimed purpose or may include a general-purpose computer selectively activated or reconfigured by a computer program stored in a computer.

[0385] Furthermore, the language used in this specification has been primarily chosen for readability and instructional purposes and may not have been chosen to depict or limit the disclosed subject matter. Therefore, the embodiments disclosed herein are intended to illustrate, and not limit, the scope of the concepts discussed herein.

Claims

1. A wallpaper display method, applied to a terminal device, characterized in that, The method comprises: detecting a first input operation of a user, wherein the first input operation is used to specify a first image as a generation source of a wallpaper, the first image comprising a first character with a first facial feature, the first character having a first display state in the first image, the first display state comprising at least one of clothing, makeup, a wearable device, a background, and a style; displaying a second image as a wallpaper, wherein the second image is generated based on the first input operation and the first image, the second image comprising the first character with the first facial feature, the first character having a second display state in the second image, the second display state comprising at least one of clothing, makeup, a wearable device, a background, and a style, and the first display state and the second display state being different.

2. The method of claim 1, wherein, The first input operation is used to specify the second display state.

3. The method of claim 2, wherein, The second display state of the first character in the second image comprises a first partial state and a second partial state, and The second image is generated based on the first input operation and the first image, comprising: The second image is obtained by processing the first image according to the first partial state and the second partial state, wherein The first partial state is determined based on the first input operation, and the second partial state is confirmed based on a default setting.

4. The method according to any one of claims 1 to 3, characterized in that, The first display state and the second display state are different, comprising: The first style of the first character in the first image is different from the second style of the first character in the second image.

5. The method of claim 4, wherein, The first input operation is used to specify the second style in the second display state, and The second image is generated based on the first input operation and the first image, comprising: The second image is obtained by processing the first image according to prompt information corresponding to the second style, wherein The prompt information corresponding to the second style is generated based on a style parameter corresponding to the second style.

6. The method according to any one of claims 1 to 5, characterized in that, The first display state further comprises a posture of the first character, and The first display state and the second display state are different, comprising: The first posture of the first character in the first image is different from the second posture of the first character in the second image.

7. The method of claim 6, wherein, The first input operation is used to specify the second posture in the second display state, and The second image is generated based on the first input operation and the first image, further comprising: The second image is obtained by processing the first character in the first image according to a posture feature vector corresponding to the second posture, wherein The posture feature vector corresponding to the second posture is obtained based on a first posture template; or The posture feature vector corresponding to the second posture is obtained by extracting feature information based on a third image specified by a user.

8. The method according to any one of claims 1 to 7, characterized in that, The method further comprises: sending image data corresponding to the first image to a server; receiving the second image returned by the server, and displaying the second image as a wallpaper.

9. The method according to any one of claims 1 to 7, characterized in that, The method comprises: displaying a third image as wallpaper, wherein the third image is obtained by adding an identification of a commodity associated with a second display state to the second image.

10. The method of claim 9, wherein, The third image is displayed as wallpaper, comprising: in response to a second input operation of a user, switching the displayed wallpaper from the second image to the third image.

11. The method of claim 10, wherein, The second input operation comprises an operation of the user clicking or circling at least one wallpaper element in the second image.

12. The method of claim 9, wherein, The method further comprises: in response to the operation of the user on the identification of the commodity, displaying a recommendation interface of the commodity.

13. The method of claim 12, wherein, The recommendation interface of the commodity is an interface of a first application installed on the terminal device.

14. The method according to any one of claims 9 to 13, characterized in that, The commodity comprises one or more of the following: corresponding to the clothing-related wallpaper element, the commodity comprises any of a hat, a piece of clothing, a pair of shoes, a necklace, a bracelet, and ear ornaments; corresponding to the makeup-related wallpaper element, the commodity comprises any of an eyebrow pencil, an eye shadow, a lipstick, a toner, and a cream; corresponding to the wearable device-related wallpaper element, the commodity comprises any of a watch, earphones, and glasses; corresponding to the background-related wallpaper element, the commodity comprises a tourism product related to a scenic spot corresponding to the background.

15. The method according to any one of claims 1 to 14, characterized in that, The first input operation is an operation of the user setting wallpaper; or The first input operation is an interface operation of a second application installed on the terminal device by the user.

16. The method according to any one of claims 1 to 14, characterized in that, The method further comprises: recommending the generated second image to a third application.

17. A wallpaper display method characterized by comprising: Comprise: The terminal device detects a first input operation of a user, wherein the first input operation is used to specify a first image as a generation source of wallpaper, the first image comprises a first character with a first facial feature, and the first character has a first display state in the first image, the first display state comprises at least one of clothing, makeup, a wearable device, a background, and a style; The terminal device sends operation data corresponding to the first input operation and the first image to a server; The server generates a second image based on the operation data corresponding to the first input operation and the first image, wherein the second image comprises the first character with the first facial feature, the first character has a second display state in the second image, the second display state comprises at least one of clothing, makeup, a wearable device, a background, and a style, and the first display state and the second display state are different; The terminal device receives the second image returned by the server and displays the second image as wallpaper.

18. A terminal device, comprising: Comprise: one or more processors; one or more memories; the one or more memories store one or more programs, when the one or more programs are executed by the one or more processors, the terminal device executes the wallpaper display method in any one of claims 1 to 17.

19. A computer readable medium characterized by The readable medium stores instructions, when the instructions are executed on a computer, the computer executes the wallpaper display method in any one of claims 1 to 17.

20. A computer program product, characterised in that, A computer program product including a computer program / instructions that, when executed by a processor, implement the wallpaper display method of any one of claims 1 to 17.

Images