Virtual avatar interaction methods, devices, electronic devices, media and program products
By acquiring images of target individuals from multiple user accounts and generating multi-person virtual avatars that match their movements and postures, the problem of lacking personalization in virtual avatars is solved, and the realism and interactivity of the virtual avatars' movements are improved.
Patent Information
- Application Number
- CN202611215278.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-08-11
- Publication Date
- 2026-11-17
AI Technical Summary
Virtual avatars on existing social platforms lack personalization, resulting in poor interactivity.
By acquiring target character images from multiple user accounts, multi-person virtual avatars are generated using a preset generation model. The actions and postures of each virtual avatar are matched one-to-one with the actions and postures of the corresponding user account in the target character images. Style options and custom style controls are provided to enhance personalization.
It enhances the realism and personalization of multi-person virtual avatars and improves the interactivity based on virtual avatars.
Smart Images

Figure CN123114342A_ABST
Abstract
Description
Technical Field
[0001] This disclosure relates to the field of computer vision technology, and in particular to a virtual avatar interaction method, device, electronic device, medium, and program product. Background Technology
[0002] Currently, some social media platforms support the creation of virtual avatars for users and interaction based on these avatars. However, in these technologies, virtual avatars are often selected by users from pre-set styles on the platform, resulting in severe homogenization and a lack of personalization, which in turn leads to problems such as poor interactivity based on virtual avatars. Summary of the Invention
[0003] This disclosure provides a method, apparatus, electronic device, medium, and program product for virtual avatar interaction, which can enhance the realism and personalization of the actions of multiple virtual avatars, and thus better improve the interactivity based on virtual avatars. The technical solution of this disclosure is as follows: According to a first aspect of the present disclosure, a virtual avatar interaction method is provided, comprising: Displays a virtual avatar interaction page shared by multiple user accounts, the virtual avatar interaction page including a preset avatar customization entry; In response to a character image acquisition instruction triggered based on the preset image customization entry, a target character image is acquired. The target character image is either a character image including the corresponding character of each of the multiple user accounts, or a character image corresponding to each of the multiple user accounts. In response to a virtual avatar generation instruction triggered based on the target person image, a first multi-person virtual avatar corresponding to the multiple user accounts is generated, and the action posture of each virtual avatar in the first multi-person virtual avatar corresponds one-to-one with the action posture of the corresponding user account in the target person image.
[0004] In an optional embodiment, the first multi-person virtual avatar is generated by a preset generation model based on the target person image and the human body movement and posture features corresponding to the user accounts in the target person image, wherein the human body movement and posture features are obtained by human body posture recognition of the target person image.
[0005] In an optional embodiment, the preset generation model includes an appearance feature extraction layer, multiple appearance action learning layers, and a virtual avatar generation layer; the appearance feature extraction layer is used to extract the appearance features corresponding to each of the multiple user accounts from the target person image; each appearance action learning layer is used to learn the appearance action based on the appearance feature corresponding to one user account and the human body action posture feature corresponding to the user account, to obtain the appearance action feature corresponding to the user account; the virtual avatar generation layer is used to generate the first multi-person virtual avatar based on the appearance action features corresponding to each of the multiple user accounts.
[0006] In an optional embodiment, before generating the first multi-user virtual avatars corresponding to the multiple user accounts in response to a virtual avatar generation instruction triggered based on the target person image, the method further includes: Display at least one style option and / or a custom style control; In response to a style setting instruction triggered based on any of the at least one style option or the custom style control, the target image style is determined; The step of generating a first multi-user virtual avatar corresponding to the multiple user accounts in response to a virtual avatar generation instruction triggered based on the target person image includes: In response to the virtual avatar generation instruction, the first multi-person virtual avatar corresponding to the target avatar style is generated.
[0007] In an optional embodiment, the custom style control is a custom style input control; displaying at least one style option and / or a custom style control includes: Display at least one style option and / or include the custom style input control, wherein the custom style input control includes reference style description information; The reference style description information is style description information extracted based on the target person image.
[0008] In an optional embodiment, after generating a first multi-user virtual avatar corresponding to the plurality of user accounts in response to a virtual avatar generation instruction triggered based on the target person image, the method further includes: The dialog pages corresponding to the multiple user accounts display the generation prompt information for the first multi-person virtual avatar and the viewing entry for the first multi-person virtual avatar.
[0009] In an optional embodiment, generating a first multi-user virtual avatar corresponding to the multiple user accounts in response to a virtual avatar generation instruction triggered based on the target person image includes: In response to the virtual avatar generation instruction, the first multi-person virtual avatar is generated when the human body action posture features corresponding to the multiple user accounts all meet the preset action constraint conditions. The preset action constraints include action confidence constraints and / or action specification constraints.
[0010] In an optional embodiment, the method further includes: In response to the virtual avatar generation instruction, if the human body posture features corresponding to any of the multiple user accounts do not meet the preset action constraint conditions, a second multi-person virtual avatar corresponding to the multiple user accounts is generated. In this context, the action posture of the first virtual image in the second multi-person virtual avatar corresponds one-to-one with the action posture of the corresponding user account in the target character image; the action posture of the second virtual image in the second multi-person virtual avatar is the corresponding preset action posture; the first virtual image is a virtual image whose human body action posture features of the corresponding user account in the second multi-person virtual avatar satisfy the preset action constraint conditions; and the second virtual image is a virtual image whose human body action posture features of the corresponding user account in the second multi-person virtual avatar do not satisfy the preset action constraint conditions.
[0011] In an optional embodiment, generating a first multi-user virtual avatar corresponding to the multiple user accounts in response to a virtual avatar generation instruction triggered based on the target person image includes: In response to the virtual avatar generation instruction, the first multi-person virtual avatar is displayed on the generation page; The method further includes: In response to the action posture update instruction corresponding to the third virtual image in the first multi-person virtual image, the first multi-person virtual image in the generated page is updated to the third multi-person virtual image corresponding to the multiple user accounts; The third virtual image is the virtual image whose action posture is updated by the action posture update instruction in the first multi-person virtual image; the third multi-person virtual image is the multi-person virtual image after updating the action posture corresponding to the third virtual image in the first multi-person virtual image to the action posture indicated by the action posture update instruction.
[0012] In an optional embodiment, displaying the first multi-person virtual avatar on the generation page in response to the virtual avatar generation instruction includes: The generated page displays the first multi-person virtual avatar and the first action description text corresponding to each virtual avatar in the first multi-person virtual avatar; In response to a first update instruction for the action description text corresponding to the third virtual character in the first action description text, the first action description text is updated; The step of updating the first multi-person virtual avatar in the generated page to the third multi-person virtual avatar corresponding to the multiple user accounts in response to the action posture update instruction for the third multi-person virtual avatar in the first multi-person virtual avatar includes: In response to the action posture update instruction triggered based on the updated first action description text, the first multi-person virtual avatar in the generated page is updated to the third multi-person virtual avatar.
[0013] In an optional embodiment, updating the first multi-person virtual avatar in the generated page to the third multi-person virtual avatar corresponding to the multiple user accounts in response to the action posture update instruction corresponding to the third multi-person virtual avatar in the first multi-person virtual avatar includes: If a movement operation is detected targeting the target part of the third virtual avatar in the first multi-person virtual avatar, the action posture update instruction is triggered to update the first multi-person virtual avatar in the generated page to the third multi-person virtual avatar.
[0014] In an optional embodiment, before generating the first multi-user virtual avatar corresponding to the plurality of user accounts in response to a virtual avatar generation instruction triggered based on the target person image, the method further includes: Display the second action description information of the multiple figures in the target figure image; In response to a second update instruction for the action description text corresponding to the target character in the second action description text, the second action description text is updated; The step of generating a first multi-user virtual avatar corresponding to the multiple user accounts in response to a virtual avatar generation instruction triggered based on the target person image includes: In response to the virtual avatar generation instruction, the first multi-person virtual avatar corresponding to the updated second action description text is generated, and the action posture of each virtual avatar in the first multi-person virtual avatar corresponds one-to-one with the action posture of the corresponding user account described in the updated second action description text.
[0015] In an optional embodiment, the method further includes: Display the first multi-person virtual avatar and at least one historical interaction information to present the interaction between corresponding user accounts based on at least some of the virtual avatars in the first multi-person virtual avatar; Wherein, the at least one historical interaction information is the historical interaction information between the user accounts corresponding to the at least some virtual images.
[0016] In an optional embodiment, after generating a first multi-user virtual avatar corresponding to the plurality of user accounts in response to a virtual avatar generation instruction triggered based on the target person image, the method further includes: In response to a dialogue interaction command triggered by a fourth virtual avatar, target dialogue information corresponding to the dialogue interaction command is sent to at least one other user account, so that when the at least one other user account views the first multi-person virtual avatar, the target dialogue information is presented in the form of a dialogue between the fourth virtual avatar and the virtual avatar corresponding to the at least one other user account. The fourth virtual avatar is the virtual avatar corresponding to the current account among the first multiple virtual avatars; the at least one other user account is a user account other than the current account among the multiple user accounts.
[0017] In an optional embodiment, the method further includes: When the interaction information between the multiple user accounts meets the preset interaction conditions, the virtual avatar interaction entry and customized unlock prompt information are displayed on the dialogue page corresponding to the multiple user accounts. The customized unlock prompt information is used to indicate that the virtual avatar customization function has been successfully unlocked. The interactive page displaying the shared virtual avatars of multiple user accounts includes: In response to a page display instruction triggered by the virtual avatar interaction entry, the virtual avatar interaction page, including the preset avatar customization entry, is displayed.
[0018] According to a second aspect of the present disclosure, a virtual avatar interaction device is provided, comprising: The page display module is configured to execute and display a virtual avatar interaction page shared by multiple user accounts, the virtual avatar interaction page including a preset avatar customization entry; The character image acquisition module is configured to execute a character image acquisition instruction triggered based on the preset image customization entry, and acquire a target character image, wherein the target character image is a character image including the character corresponding to each of the multiple user accounts, or is a character image corresponding to each of the multiple user accounts. The first image generation module is configured to execute a virtual image generation instruction triggered based on the target person image to generate a first multi-person virtual image corresponding to the multiple user accounts, and the action posture of each virtual image in the first multi-person virtual image corresponds one-to-one with the action posture of the corresponding user account in the target person image.
[0019] In an optional embodiment, the first multi-person virtual avatar is generated by a preset generation model based on the target person image and the human body movement and posture features corresponding to the user accounts in the target person image, wherein the human body movement and posture features are obtained by human body posture recognition of the target person image.
[0020] In an optional embodiment, the preset generation model includes an appearance feature extraction layer, multiple appearance action learning layers, and a virtual avatar generation layer; the appearance feature extraction layer is used to extract the appearance features corresponding to each of the multiple user accounts from the target person image; each appearance action learning layer is used to learn the appearance action based on the appearance feature corresponding to one user account and the human body action posture feature corresponding to the user account, to obtain the appearance action feature corresponding to the user account; the virtual avatar generation layer is used to generate the first multi-person virtual avatar based on the appearance action features corresponding to each of the multiple user accounts.
[0021] In an optional embodiment, the apparatus further includes: The style information display module is configured to display at least one style option and / or a custom style control; The style determination module is configured to execute a style setting instruction triggered in response to any of the at least one style option or the custom style control to determine the target image style; The first image generation module includes: The first image generation unit is configured to execute a virtual image generation instruction in response to generate the first multi-person virtual image corresponding to the target image style.
[0022] In an optional embodiment, the style information display module is further configured to display at least one style option and / or include the custom style input control, wherein the custom style input control includes reference style description information; The reference style description information is style description information extracted based on the target person image.
[0023] In an optional embodiment, the apparatus further includes: The prompt generation module is configured to display the prompt information corresponding to the first multi-person virtual avatar and the viewing entry corresponding to the first multi-person virtual avatar on the dialogue pages corresponding to the multiple user accounts.
[0024] In an optional embodiment, the first image generation module includes: The second image generation unit is configured to execute a response to the virtual image generation instruction and generate the first multi-person virtual image when the human body action posture features corresponding to the multiple user accounts all meet the preset action constraint conditions. The preset action constraints include action confidence constraints and / or action specification constraints.
[0025] In an optional embodiment, the apparatus further includes: The second image generation module is configured to execute a response to the virtual image generation instruction, and generate a second multi-person virtual image corresponding to the multiple user accounts when the human body action posture features corresponding to any user account among the multiple user accounts do not meet the preset action constraint conditions. In this context, the action posture of the first virtual image in the second multi-person virtual avatar corresponds one-to-one with the action posture of the corresponding user account in the target character image; the action posture of the second virtual image in the second multi-person virtual avatar is the corresponding preset action posture; the first virtual image is a virtual image whose human body action posture features of the corresponding user account in the second multi-person virtual avatar satisfy the preset action constraint conditions; and the second virtual image is a virtual image whose human body action posture features of the corresponding user account in the second multi-person virtual avatar do not satisfy the preset action constraint conditions.
[0026] In an optional embodiment, the first image generation module includes: The virtual avatar display module is configured to respond to the virtual avatar generation instruction and display the first multi-person virtual avatar on the generation page; The device further includes: The virtual avatar update unit is configured to execute an action posture update instruction in response to the third virtual avatar in the first multi-person virtual avatar, and update the first multi-person virtual avatar in the generated page to the third multi-person virtual avatar corresponding to the multiple user accounts. The third virtual image is the virtual image whose action posture is updated by the action posture update instruction in the first multi-person virtual image; the third multi-person virtual image is the multi-person virtual image after updating the action posture corresponding to the third virtual image in the first multi-person virtual image to the action posture indicated by the action posture update instruction.
[0027] In an optional embodiment, the virtual avatar display module includes: The first information display unit is configured to display the first multi-person virtual avatar and the first action description text corresponding to each virtual avatar in the first multi-person virtual avatar on the generated page. The action update unit is configured to execute an update instruction in response to a first action description text corresponding to the third virtual image in the first action description text, and update the first action description text. The virtual avatar update unit includes: The first update unit is configured to execute an action posture update instruction triggered based on the updated first action description text, and update the first multi-person virtual avatar in the generated page to the third multi-person virtual avatar.
[0028] In an optional embodiment, the virtual avatar updating unit includes: The second update unit is configured to, upon detecting a movement operation on the target part corresponding to the third virtual image in the first multi-person virtual image, trigger the action posture update instruction to update the first multi-person virtual image in the generated page to the third multi-person virtual image.
[0029] In an optional embodiment, the apparatus further includes: The descriptive text display module is configured to display second action description information of the multiple figures in the target figure image; The action update module is configured to execute a second update instruction in response to the action description text corresponding to the target character in the second action description text, and update the second action description text. The first image generation module includes: The third image generation unit is configured to execute a response to the virtual image generation instruction to generate the first multi-person virtual image corresponding to the updated second action description text, wherein the action posture of each virtual image in the first multi-person virtual image corresponds one-to-one with the character action posture of the corresponding user account described in the updated second action description text.
[0030] In an optional embodiment, the apparatus further includes: The historical interaction module is configured to display the first multi-person virtual avatars and at least one historical interaction information to present the interaction between corresponding user accounts based on at least some of the virtual avatars in the first multi-person virtual avatars. Wherein, the at least one historical interaction information is the historical interaction information between the user accounts corresponding to the at least some virtual images.
[0031] In an optional embodiment, the apparatus further includes: The dialogue interaction module is configured to execute a dialogue interaction command triggered based on the fourth virtual avatar, and send the target dialogue information corresponding to the dialogue interaction command to at least one other user account, so that when the at least one other user account views the first multi-person virtual avatar, the target dialogue information is presented in the form of a dialogue between the fourth virtual avatar and the virtual avatar corresponding to the at least one other user account. The fourth virtual avatar is the virtual avatar corresponding to the current account among the first multiple virtual avatars; the at least one other user account is a user account other than the current account among the multiple user accounts.
[0032] In an optional embodiment, the apparatus further includes: The avatar customization prompt module is configured to display a virtual avatar interaction entry and customization unlock prompt information on the dialogue page corresponding to the multiple user accounts when the interaction information between the multiple user accounts meets the preset interaction conditions. The customization unlock prompt information is used to indicate that the virtual avatar customization function has been successfully unlocked. The page display module includes: The page display unit is configured to execute a page display instruction triggered based on the virtual avatar interaction entry, and display the virtual avatar interaction page including the preset avatar customization entry.
[0033] According to a third aspect of the present disclosure, an electronic device is provided, comprising: a processor; and a memory for storing processor-executable instructions; wherein the processor is configured to execute the instructions to implement any virtual avatar interaction method provided in the present disclosure.
[0034] According to a fourth aspect of the present disclosure, a computer-readable storage medium is provided, wherein when instructions in the storage medium are executed by a processor of an electronic device, the electronic device is enabled to perform any virtual avatar interaction method provided in the present disclosure. According to a fifth aspect of the present disclosure, a computer program product comprising computer program instructions is provided, which, when executed by a processor on a computer, cause the computer to perform any of the virtual avatar interaction methods provided in the present disclosure as described above.
[0035] The technical solutions provided by the embodiments of this disclosure have at least the following beneficial effects: This page displays a shared virtual avatar interaction page for multiple user accounts. The virtual avatar interaction page includes a preset avatar customization entry point. In response to a character image acquisition command triggered by the preset avatar customization entry point, a target character image is acquired. This target character image includes character images corresponding to each of the multiple user accounts, or it is a single character image corresponding to each of the multiple user accounts. In response to a virtual avatar generation command triggered by the target character image, a first multi-user virtual avatar corresponding to the multiple user accounts is generated. The action posture of each virtual avatar in the first multi-user virtual avatar corresponds one-to-one with the action posture of the corresponding user account in the target character image. Using the technical solution provided in this embodiment, personalized customization of multi-user virtual avatars corresponding to multiple accounts can be achieved. Furthermore, during the personalized customization process, by acquiring character images corresponding to each of the multiple user accounts, or by acquiring character images corresponding to each of the multiple user accounts including multiple user accounts, the action posture of each virtual avatar in the multi-user virtual avatar can be personalized based on the action posture of the corresponding user account in the character image. This restores the real action posture of multiple users, better enhancing the realism and personalization of the multi-user virtual avatar's actions, and thus also improving the interactivity based on the virtual avatar.
[0036] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description
[0037] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure, and are not intended to unduly limit this disclosure.
[0038] Figure 1 This is a schematic diagram illustrating an application environment according to an exemplary embodiment; Figure 2 This is a flowchart illustrating a virtual avatar interaction method according to an exemplary embodiment; Figure 3 This is a schematic diagram of a chat page corresponding to multiple user accounts, according to an exemplary embodiment. Figure 4 This is a schematic diagram illustrating page changes during the acquisition of a target person image according to an exemplary embodiment; Figure 5 This is a schematic diagram of a dialog page that displays a prompt message for generating a first multi-person virtual avatar and an entry point for viewing the first multi-person virtual avatar on different user account sides, according to an exemplary embodiment. Figure 6This is a schematic diagram of a page displaying different pages with a first set of virtual avatars of multiple people, provided according to an exemplary embodiment; Figure 7 This is a schematic diagram of a page displaying a first set of virtual avatars of multiple people and at least one set of historical interaction information, according to an exemplary embodiment. Figure 8 This is a schematic diagram illustrating, according to an exemplary embodiment, a form in which target dialogue information is presented in the form of a fourth virtual avatar engaging in dialogue with at least one other user account's corresponding virtual avatar; Figure 9 This is a block diagram illustrating a virtual avatar interaction device according to an exemplary embodiment; Figure 10 This is a block diagram illustrating an electronic device for interacting with virtual avatars according to an exemplary embodiment. Detailed Implementation
[0039] To enable those skilled in the art to better understand the technical solutions of this disclosure, the technical solutions in the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings.
[0040] It should be noted that the terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this disclosure are used to distinguish similar concepts and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this disclosure described herein can be implemented in orders other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this disclosure. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this disclosure as detailed in the appended claims.
[0041] It should be noted that the user information (including but not limited to user device information, user images and other personal information) and data (including but not limited to data used for display and analysis) involved in this disclosure are all information and data authorized by the user or fully authorized by all parties.
[0042] Please see Figure 1 , Figure 1 This is a schematic diagram illustrating an application environment according to an exemplary embodiment, which may include a terminal 100 and a server 300.
[0043] In an optional embodiment, the first terminal 100 can be a terminal corresponding to any one of multiple user accounts among multiple user accounts for which a multi-person virtual avatar is generated. Each virtual avatar in the multi-person virtual avatar corresponds to one of the multiple user accounts. The multi-person virtual avatar can be a personalized multi-person virtual avatar customized and generated by any one of the multiple user accounts. Optionally, if no corresponding personalized multi-person virtual avatar is generated by multiple user accounts, the multi-person virtual avatar can be a platform preset or a multi-person virtual avatar selected by any user account from multiple preset outfit styles provided by the platform. Specifically, the terminal 100 can be, but is not limited to, electronic devices such as smartphones, desktop computers, tablets, laptops, smart speakers, digital assistants, augmented reality (AR) / virtual reality (VR) devices, and smart wearable devices. It can also be software running on the aforementioned electronic devices, such as applications. Optionally, the operating system running on the electronic device can be, but is not limited to, Android, iOS, Linux, and Windows.
[0044] In an optional embodiment, server 200 can provide background services to terminal 100. Specifically, server 200 can be a standalone physical server, a server cluster or distributed system consisting of multiple physical servers, or a cloud server providing cloud computing services.
[0045] In addition, it should be noted that, Figure 1 The example shown is merely one application environment provided by this disclosure. In practical applications, other application environments may also be included, such as more terminals.
[0046] In the embodiments described in this specification, the terminal 100 and the server 200 can be directly or indirectly connected through wired or wireless communication, and this disclosure does not impose any restrictions.
[0047] Figure 2 This is a flowchart illustrating a virtual avatar interaction method according to an exemplary embodiment. This method can be applied to a terminal, such as... Figure 2 As shown, the method may include the following steps: In step S201, a virtual avatar interaction page shared by multiple user accounts is displayed. The virtual avatar interaction page includes a preset avatar customization entry.
[0048] In one specific embodiment, the virtual avatar interaction page can be used to dress up a shared multi-person virtual avatar for multiple user accounts; optionally, dressing up a multi-person virtual avatar can include customizing a personalized multi-person virtual avatar, that is, generating a multi-person virtual avatar (personalized virtual avatar) based on a character image provided by any user account among multiple user accounts; it can also include dressing up a virtual avatar based on preset dressing styles, that is, dressing up a multi-person virtual avatar by selecting any preset dressing style from at least one preset dressing style.
[0049] In an optional embodiment, if none of the multiple user accounts have customized corresponding personalized multi-person virtual avatars, the multi-person virtual avatars on the virtual avatar interaction page can be preset by the platform, or they can be a multi-person virtual avatar selected by any user account from multiple preset avatar styles provided by the platform. The current account can be any one of the multiple user accounts; that is, any one of the multiple user accounts can customize corresponding personalized multi-person virtual avatars for multiple user accounts. The multiple user accounts can be at least two user accounts.
[0050] In an optional embodiment, multiple user accounts can be pre-bound with preset relationships. After the relationship binding is completed, the platform will provide exclusive virtual interaction spaces for multiple user accounts and load corresponding multi-person virtual avatars on the virtual interaction space page (landing page) for interaction based on the multi-person virtual avatars. Specifically, the preset relationships can be set according to actual needs, such as family relationships, romantic relationships, best friend relationships, etc. Optionally, the interaction based on the multi-person virtual avatars can include the above-mentioned dressing-up interaction for the multi-person virtual avatars, and can also include the execution of at least one preset interactive operation. Specifically, at least one preset interactive operation can be set according to actual application needs, such as inviting other accounts to browse the virtual interaction spaces corresponding to multiple user accounts, inviting other accounts to perform interactive operations such as liking the multi-person virtual avatars corresponding to multiple user accounts, and interactive operations such as jointly raising virtual pets; optionally, performing at least one preset interactive operation can obtain corresponding virtual currency (such as digital currency or platform-wide virtual currency that complies with laws and regulations).
[0051] In an optional embodiment, an entry point for the virtual avatar interaction page can be set on a chat page corresponding to multiple user accounts or a virtual interaction space page shared by multiple user accounts to trigger the display of the virtual avatar interaction page. Optionally, the personalized virtual avatar customization function can be automatically enabled for user accounts bound to a preset relationship, or it can be enabled only when the interaction information between user accounts bound to a preset relationship meets preset interaction conditions. Optionally, the above method may further include: When the interaction information between multiple user accounts meets the preset interaction conditions, the virtual avatar interaction entry and customized unlock prompt information are displayed on the dialogue page corresponding to the multiple user accounts; Accordingly, the aforementioned interactive page displaying shared virtual avatars from multiple user accounts may include: In response to a page display command triggered by the virtual avatar interaction portal, a virtual avatar interaction page, including a preset avatar customization portal, is displayed.
[0052] In one specific embodiment, the interaction information between multiple user accounts may include dialogue interaction information between multiple user accounts, and may also include work interaction information (information on at least one interactive operation such as interactive comments, likes, or shares for each user's published work), live streaming interaction information (interaction information for each user's live stream and / or interaction information for joint live streams), etc. Specifically, the preset interaction conditions may be pre-set interaction conditions for unlocking the virtual avatar customization function; for example, preset interaction number, preset consecutive interaction days, etc. Correspondingly, if the interaction information between multiple user accounts indicates that the number of interactions between multiple user accounts is greater than the preset interaction number, or if the interaction information between multiple user accounts indicates that the consecutive interaction days between multiple user accounts are greater than the preset consecutive interaction days, it can be determined that the interaction information between multiple user accounts meets the preset interaction conditions.
[0053] In one specific embodiment, the customized unlock prompt message can be used to indicate that the virtual avatar customization function (the function of personalized virtual avatar customization) has been successfully unlocked (i.e., the virtual avatar customization function has been activated); the virtual avatar interaction entry can be the entry that triggers the display of the virtual avatar interaction page; optionally, the above-mentioned page display instruction can be triggered by touch operations such as clicking or long-pressing the virtual avatar interaction entry to display the virtual avatar interaction page including the preset avatar customization entry.
[0054] In a specific embodiment, such as Figure 3 As shown, Figure 3 This is a schematic diagram of a dialogue page corresponding to multiple user accounts according to an exemplary embodiment; wherein, the information corresponding to 301 can be a virtual avatar interaction entry, and the information corresponding to 302 can be customized unlock prompt information.
[0055] In the above embodiments, when the interaction information between multiple user accounts meets the preset interaction conditions, while displaying the virtual avatar interaction entry on the dialogue page corresponding to the multiple user accounts, a customization unlock prompt message indicating that the virtual avatar customization function has been successfully unlocked is also displayed. This makes it easier for users to understand that the virtual avatar customization function has been unlocked, thereby increasing users' enthusiasm for entering the virtual avatar interaction page to customize their personal virtual avatars and improving the convenience of virtual avatar customization interaction.
[0056] In one specific embodiment, the preset avatar customization entry can be an entry used to trigger personalized customization of virtual avatars for multiple user accounts. This preset avatar customization entry can be a touch-sensitive control.
[0057] In an optional embodiment, a preset image customization entry can be set directly on the aforementioned virtual interactive space page. Correspondingly, the aforementioned virtual image interaction page can also be a virtual interactive space page.
[0058] In step S203, in response to the character image acquisition instruction triggered based on the preset image customization entry, the target character image is acquired; In one specific embodiment, the character image acquisition instruction can be an instruction to acquire a character image used for virtual avatar generation (i.e., personalized virtual avatar customization). Optionally, the character image setting control can be displayed first based on a preset avatar customization entry point, so as to acquire and set the target character image based on the character image setting control. Optionally, the acquisition of the target character image in response to the character image acquisition instruction triggered based on the preset avatar customization entry point can include: in response to a customization trigger instruction triggered based on the preset avatar customization entry point, displaying a virtual avatar customization page including character image setting controls; and in response to the character image acquisition instruction triggered by the character image setting control, displaying the acquired target character image on the virtual avatar customization page.
[0059] In one specific embodiment, the customization trigger instruction can be an instruction to begin customizing a virtual avatar; optionally, it can be a customization trigger instruction triggered by an operation such as clicking a preset avatar customization entry. Specifically, the virtual avatar customization page can be a page for personalized customization of a virtual avatar; optionally, the virtual avatar customization page can be a new page or the aforementioned virtual avatar interaction page. For example, when the customization trigger instruction is triggered based on a preset avatar customization entry, a character image setting control is displayed on the virtual avatar interaction page to set the target character image based on the character image setting control. Specifically, the character image setting control can be used to set the target character image. Optionally, the target character image can be set by uploading, real-time shooting, or selecting account avatars (including character account avatars) from multiple user accounts; correspondingly, the character image setting control can include at least one of an image upload control or an image selection control.
[0060] In one specific embodiment, the target person image can be a person image including the person corresponding to each of multiple user accounts (i.e., a group photo of multiple people), or a person image corresponding to each of multiple user accounts (i.e., individual person images of multiple people); the target person image can be a real person image corresponding to multiple user accounts.
[0061] In a specific embodiment, such as Figure 4 As shown, Figure 4 This is a schematic diagram illustrating page changes during the acquisition of a target person image according to an exemplary embodiment. Specifically, Figure 4 Figure 'a' shows a virtual avatar interaction page with a preset avatar customization entry point, and the control corresponding to 401 can be the preset avatar customization entry point; optionally, clicking the preset avatar customization entry point or other operations can trigger... Figure 4 The virtual avatar customization page shown in b includes controls for setting the character image, and the control corresponding to 402 can be a character image setting control; furthermore, combined with Figure 4 As shown in Figure c, the target person image 403 can be set (uploaded) based on the person image setting control. Optionally, the target person image 403 can be a group photo image corresponding to multiple user accounts. If the first uploaded person image is a single person image, the person image setting control can be displayed while displaying the person image to continue uploading person images.
[0062] In step S205, in response to the virtual avatar generation instruction triggered based on the target person image, a first multi-person virtual avatar corresponding to multiple user accounts is generated, and the action posture of each virtual avatar in the first multi-person virtual avatar corresponds one-to-one with the action posture of the corresponding user account in the target person image.
[0063] In one specific embodiment, the virtual avatar generation instruction can be an instruction to generate multiple virtual avatars corresponding to multiple user accounts based on a target person image. The first multi-person virtual avatar may include the virtual avatars corresponding to each of the multiple user accounts; specifically, the aforementioned first multi-person virtual avatar is generated by a preset generation model based on the target person image and the human motion posture features corresponding to each user account in the target person image, and the human motion posture features are obtained by performing human posture recognition on the target person image.
[0064] In one specific embodiment, the preset generation model can be a generative intelligent model used for generating virtual avatars. Specifically, the human motion posture features corresponding to each user account (each character) can be information that can characterize the corresponding motion posture of the character in the target character image. For example, it can include a set of human keypoint coordinates of the character in the corresponding motion posture (e.g., the set of human keypoint coordinates of two people with their arms around each other) or a motion posture feature vector characterizing the corresponding motion posture of the character (e.g., the motion posture feature vector of a certain dance posture during a dance). Optionally, artificial intelligence technologies such as skeleton keypoint detection, depth map pose estimation, optical flow pose extraction, and multi-view pose fusion can be used to identify the human motion posture features corresponding to each character.
[0065] In the above embodiments, the human body movement and posture features corresponding to each user account are first identified from the target person image, and then input together with the person image into the preset generation model. This allows the model to better learn the movement features of each person, improve the effect of the generated multi-person virtual image on providing users with the restoration of real human movement and posture, and thus better improve the realism and personalization of the virtual image's movements.
[0066] In one specific embodiment, the model structure of the preset generation model can be set according to the actual application requirements, such as a diffusion model; optionally, the preset generation model includes an appearance feature extraction layer, multiple appearance action learning layers and a virtual image generation layer.
[0067] In one specific embodiment, the appearance feature extraction layer can be used to extract the appearance features corresponding to multiple user accounts from the target person image. The layer structure of the appearance feature extraction layer can be set according to actual application requirements. The appearance features corresponding to each user account can represent the appearance features of the person corresponding to that user account in the target person image, and can also represent the style features of the person, such as style features extracted in combination with the person's clothing. In a specific embodiment, each appearance action learning layer can be used to learn the appearance action of a person based on the appearance features of a person corresponding to one user account and the human body action posture features of a person corresponding to one user account among multiple user accounts, so as to obtain the appearance action features of a person corresponding to a user account; specifically, the layer structure of the appearance action learning layer can be set according to the actual application requirements, and the appearance action features of a person corresponding to each user account can represent the appearance features and action posture features of the person corresponding to the user account in the target person image.
[0068] In one specific embodiment, the virtual avatar generation layer is used to generate a first multi-person virtual avatar based on the appearance and action features of the characters corresponding to multiple user accounts. The layer structure of the virtual avatar generation layer can be set according to actual application requirements.
[0069] In one specific embodiment, the target person image can first be input into the person appearance feature extraction layer for appearance feature extraction processing to obtain the appearance features corresponding to each user account (person) in the target person image; optionally, when the target person image includes person images corresponding to multiple user accounts, person detection can be performed on the target person image first, and person region images corresponding to each person can be obtained. The person region images corresponding to each person can then be input into the person appearance feature extraction layer for appearance feature extraction processing to obtain the appearance features corresponding to each user account (person); next, the appearance features corresponding to each person (user account) and the human body action posture features corresponding to that person can be input into the appearance action learning layer corresponding to that person (user account) for person appearance action learning to obtain the appearance action features corresponding to each person; finally, the appearance action features corresponding to each person can be input into the virtual image generation layer for virtual image generation processing to obtain the first multi-person virtual image.
[0070] In the above embodiments, the preset generation model includes an appearance feature extraction layer, multiple appearance action learning layers, and a virtual image generation layer. It can extract the appearance features of each character separately during the generation of multiple virtual images, and learn the appearance action features of each character separately by combining their corresponding human body action posture point features. This can effectively ensure the restoration effect of the real appearance action of each character during the generation of multiple virtual images, avoid the mutual influence of the appearance action features of different characters, and better improve the realism and personalization of the virtual image's actions in the multi-virtual image generation scenario.
[0071] In a specific embodiment, when the target character image is a character image that includes characters corresponding to multiple user accounts, the relative positional relationship between the virtual characters in the first multi-person virtual avatar is consistent with the relative positional relationship between the characters in the target character image. When the target character image is a character image that corresponds to multiple user accounts, the correspondence between the characters corresponding to the virtual characters in the first multi-person virtual avatar from left to right and the multiple character images can be determined based on the upload order of the multiple character images. For example, the virtual character corresponding to the first uploaded character image can be the first virtual character from left to right in the first multi-person virtual avatar, and so on.
[0072] In an optional embodiment, generating a first multi-user virtual avatar corresponding to multiple user accounts in response to a virtual avatar generation instruction triggered based on a target person's image may include: In response to the virtual avatar generation command, a first multi-person virtual avatar is generated when the human body posture features corresponding to multiple user accounts all meet the preset action constraints.
[0073] In one specific embodiment, the preset action constraints may include action confidence constraints and / or action specification constraints; optionally, when the preset action constraints may include action confidence constraints and action specification constraints, if any human action posture feature satisfies both the action confidence constraints and the action specification constraints, it can be determined that the human action posture feature satisfies the preset action constraints.
[0074] In one specific embodiment, action confidence constraints can be used to indicate constraints on the reliability of human action posture features, such as a lower confidence limit characterizing the reliability of human action posture features. Specifically, in the process of identifying the human action posture features corresponding to each person by combining artificial intelligence technologies such as skeleton keypoint detection, confidence data characterizing the reliability of the corresponding human action posture features will also be acquired. Specifically, the corresponding confidence data can be determined by combining dimensions such as whether keypoints such as elbows and knees are occluded, whether keypoints are located at the edge of the image and are cropped, and the clarity of the texture around the keypoints. Optionally, if the confidence data corresponding to any human action posture feature is greater than or equal to the lower confidence limit, it can be determined that the human action posture feature meets the action confidence constraints. Specifically, action specification constraints can be used to indicate constraints on the standardization of human action posture features, such as a preset abnormal action posture feature set. Correspondingly, if any human action posture feature does not match any abnormal action posture feature in the preset abnormal action posture feature set, it can be determined that the human action posture feature meets the action specification constraints.
[0075] In the above embodiments, if the human body movement and posture features identified from the target person image all meet the preset action constraints, a first multi-person virtual image can be generated based on the target person image while ensuring the reliability and effectiveness of the human body movement and posture features themselves. This can also effectively ensure the accuracy of the generated multi-person virtual image in restoring the real person's movement and posture.
[0076] In an optional embodiment, the above method may further include: In response to the virtual avatar generation command, if the human body action posture features corresponding to any user account among multiple user accounts do not meet the preset action constraints, a second multi-person virtual avatar corresponding to multiple user accounts is generated. In one specific embodiment, the action posture of the first virtual image in the second multi-person virtual avatar is matched one-to-one with the action posture of the corresponding user account in the target person image, and the action posture of the second virtual image in the second multi-person virtual avatar is the corresponding preset action posture; the first virtual image can be a virtual image in the second multi-person virtual avatar whose human body action posture features of the corresponding user account meet the preset action constraint conditions; the second virtual image can be a virtual image in the second multi-person virtual avatar whose human body action posture features of the corresponding user account do not meet the preset action constraint conditions.
[0077] In a specific embodiment, the detailed process of generating a second multi-person virtual image corresponding to multiple user accounts can be referred to the detailed process of generating a first multi-person virtual image corresponding to multiple user accounts described above, and will not be repeated here. That is, in the process of generating virtual images using a preset generation model, human motion posture features that do not meet the preset motion constraints in the human motion posture features extracted from the target person image are replaced with preset motion posture features corresponding to the preset motion posture.
[0078] In one specific embodiment, the preset action posture can be an action posture pre-set by the platform, or an action posture corresponding to a randomly generated action posture feature that meets the preset action constraints, or an action posture selected by the user from the action postures pre-set by the platform.
[0079] In an optional embodiment, when generating a second multi-person virtual avatar corresponding to multiple user accounts, corresponding action adjustment prompts can be displayed while displaying the second multi-person virtual avatar, so that users can regenerate a multi-person virtual avatar that restores the real action posture.
[0080] In an optional embodiment, if the human body posture features corresponding to any of the multiple user accounts do not meet the preset action constraints, the process of generating a multi-person virtual avatar may only restore the appearance of the character without restoring the actions.
[0081] In the above embodiments, if there are human motion posture features in the human motion posture features extracted based on the target person image that do not meet the preset motion constraints, the corresponding human motion posture features can be replaced based on the human motion posture features corresponding to the preset motion posture. This ensures that even when some real motion postures cannot be restored, the user's need to generate multiple virtual avatars can still be met, thereby improving the user experience.
[0082] In an optional embodiment, before generating a first multi-user virtual avatar corresponding to multiple user accounts in response to a virtual avatar generation instruction triggered based on a target person image, the above method may further include: Display at least one style option and / or a custom style control; In response to a style setting instruction triggered based on any style option or a custom style control, determine the target image style; Accordingly, the above-mentioned response to the virtual avatar generation instruction triggered based on the target person's image, generating a first multi-user virtual avatar corresponding to multiple user accounts, may include: In response to the virtual avatar generation command, generate the first multi-person virtual avatar corresponding to the target avatar style.
[0083] In one specific embodiment, at least one style option can be an option for at least one image style; at least one image style can include at least one preset image style provided by the platform and / or the character style corresponding to the target character image; for example, Figure 4 The style option corresponding to 404 can be the style option for the target person image; the style option corresponding to 405 can be a preset image style provided by the platform. Specifically, the image style (personal style) can be set according to actual application needs, such as modern style, campus style, ancient style, elf style, etc. The person style corresponding to the target person image can be extracted from the target person image based on a preset style extraction model. The custom style control can be used to customize the image style.
[0084] In an optional embodiment, at least one style option and / or a custom style control can be displayed on the virtual avatar customization page. Specifically, the style setting instruction can be an instruction to set the image style of the multi-person virtual avatar to be generated; optionally, the style setting instruction can be triggered by clicking any style option among at least one style option; alternatively, the style setting instruction can also be triggered by editing custom style description information through a custom style control.
[0085] In an optional embodiment, the aforementioned custom style control can be a custom style input control; the aforementioned display of at least one style option and / or custom style control can include: Display at least one style option and / or include a custom style input control, wherein the custom style input control includes reference style description information; In one specific embodiment, the reference style description information can be style description information extracted based on the target person image, that is, style description text corresponding to the style of the person in the target image. Optionally, in a scenario where the virtual avatar style is customized based on a custom style control, the style described by the reference style description information can be directly used as the target avatar style, or the style indicated by the style description information after editing and updating the reference style description information can be used as the target avatar style.
[0086] In the above embodiments, while providing a custom style input control, reference style description information is given based on the style description information extracted from the target person image. This can improve the adaptability between the reference style description information and the style of the person image provided by the user, while also improving the convenience and efficiency of custom styles.
[0087] In one specific embodiment, the aforementioned first multi-person virtual avatar is generated by a target generation model corresponding to the target image style, based on the target person image and the human motion posture features corresponding to the user accounts within the target person image. The human motion posture features are obtained by performing human posture recognition on the target person image. The target generation model is trained on a preset generation model based on multiple sample person images and virtual avatars of the target image style corresponding to each sample person image. That is, after training the preset generation model, target generation models capable of generating corresponding multi-person virtual avatars can be trained for different image styles.
[0088] In an optional embodiment, the multiple sample character images can be real character images used to train the target image style correspondence target generation model, and the virtual image corresponding to the target image style of any sample character image can be the virtual image of the real character in the sample character image under the target image style.
[0089] In an optional embodiment, a LoRA (Low-Rank Adaptation) layer can be superimposed on the preset generative model. Accordingly, the above-mentioned training of the preset generative model based on multiple sample character images and virtual images of the target image style corresponding to each sample character image may include training the preset generative model after superimposing the LoRA layer based on multiple sample character images and virtual images of the target image style corresponding to each sample character image. Specifically, during the training process, the model parameters in the preset generative model can remain unchanged, and only the weights (model parameters) of the LoRA layer are finely adjusted and updated. The trained preset generative model with the LoRA layer superimposed is used as the target generative model.
[0090] In an optional embodiment, the style description information corresponding to the target image style can be input into a preset generation model, so that the preset generation model can generate a multi-person virtual image of the target image style while restoring the action posture and the appearance features of the person. Specifically, the style description information can be input into the appearance feature extraction layer of the preset generation model together with the target person image for appearance feature extraction processing.
[0091] In the above embodiments, during the personalized multi-person virtual avatar generation process, at least one style option and / or custom style control can be set, allowing users to set the virtual avatar style by selecting a style option or by custom editing, which can greatly improve the convenience and flexibility of users in setting the virtual avatar style.
[0092] In one specific embodiment, the first multi-person virtual avatar may include one or more multi-person virtual avatars. When the first multi-person virtual avatar includes multiple multi-person virtual avatars, each multi-person virtual avatar includes virtual avatars corresponding to multiple user accounts, and the action posture of each virtual avatar in each multi-person virtual avatar corresponds one-to-one with the action posture of the corresponding user account in the target person image. Furthermore, the appearance features and image style of the multiple multi-person virtual avatars can be consistent, but the details of the multiple multi-person virtual avatars are different, such as different clothing colors, backgrounds, lighting effects, etc.
[0093] In an optional embodiment, during the generation of the first multi-person virtual avatar, the user can close the current generation page. Correspondingly, after generation is complete, a corresponding generation notification message can be sent through channels such as a chat page, so that multiple user accounts can promptly view the generated first multi-person virtual avatar. Accordingly, after generating the first multi-person virtual avatar corresponding to multiple user accounts in response to a virtual avatar generation command triggered based on a target person's image, the above method may further include: The dialog pages corresponding to multiple user accounts display the generation prompt information for the first multi-person virtual avatar and the viewing entry for the first multi-person virtual avatar.
[0094] In one specific embodiment, the generated prompt message can be used to indicate that a personalized multi-person virtual avatar (first multi-person virtual avatar) corresponding to multiple user accounts has been generated; the viewing entry corresponding to the first multi-person virtual avatar can be used to trigger the viewing of the first multi-person virtual avatar.
[0095] In a specific embodiment, taking multiple user accounts as two user accounts as an example, such as... Figure 5 As shown, Figure 5 This is a schematic diagram of a page illustrating a dialogue page displaying a prompt message for generating a first multi-user virtual avatar and an entry point for viewing the first multi-user virtual avatar, according to an exemplary embodiment; wherein, Figure 5 The diagram in section a shows the dialog page on the current account side when the first multi-person virtual avatar is generated among multiple user accounts. The information corresponding to 501 can be the generation prompt information for the first multi-person virtual avatar, and the information corresponding to 502 can be the viewing entry for the first multi-person virtual avatar. Figure 5The image shown in b is the chat page of the other user account (a user account other than the current account) among the two user accounts. The information corresponding to 503 can be the generation prompt information for the first multi-person virtual avatar, and the information corresponding to 504 can be the viewing entry for the first multi-person virtual avatar.
[0096] In an optional embodiment, the above-mentioned viewing entry can trigger the display of a page showing the generation of the first multi-person virtual avatar, or it can trigger the display of a page showing the first multi-person virtual avatar, which can be set according to actual application needs.
[0097] In a specific embodiment, such as Figure 6 As shown, Figure 6 This is a schematic diagram of a page displaying different pages with a first set of virtual avatars, provided according to an exemplary embodiment; wherein, Figure 6 The image shown in section a is a schematic diagram of the page displaying the generation page with the first multi-person virtual avatars; Figure 6 The image shown in b is a schematic diagram of a virtual interactive space page displaying multiple virtual avatars.
[0098] In the above embodiments, displaying the generation prompt information and the viewing entry for the first multi-person virtual avatar on the dialogue page corresponding to multiple user accounts can facilitate users to view the generated multi-person virtual avatar in a timely manner.
[0099] In an optional embodiment, after generating the first multi-person virtual avatar, the user can adjust the action posture of at least one of the virtual avatars according to actual needs. Accordingly, the above-mentioned generation of the first multi-person virtual avatar corresponding to multiple user accounts in response to the virtual avatar generation instruction triggered based on the target person image may include: In response to the virtual avatar generation command, the first multi-person virtual avatar is displayed on the generation page; Correspondingly, the above methods may also include: In response to the action and posture update command for the third virtual avatar in the first multi-person virtual avatar, the first multi-person virtual avatar on the generated page will be updated to the third multi-person virtual avatar corresponding to multiple user accounts; In one specific embodiment, the third virtual avatar can be one or more virtual avatars whose action posture is updated by the action posture update command in the first multi-person virtual avatar; the third multi-person virtual avatar can be a multi-person virtual avatar whose action posture corresponding to the third virtual avatar in the first multi-person virtual avatar is updated to the action posture indicated by the action posture update command. Specifically, the action posture update command can be an instruction to update the action posture corresponding to the third virtual avatar in the first multi-person virtual avatar.
[0100] In an optional embodiment, the virtual avatar's corresponding action posture can be updated by adjusting the action description text. Accordingly, the above-mentioned response to the virtual avatar generation instruction, displaying the first multi-person virtual avatar on the generation page, may include: The generated page displays the first multi-person virtual avatar and the first action description text corresponding to each virtual avatar in the first multi-person virtual avatar; In response to a first update instruction for the action description text corresponding to the third virtual character in the first action description text, update the first action description text; Accordingly, the above response to the action and posture update instruction corresponding to the third virtual avatar in the first multi-person virtual avatar, updating the first multi-person virtual avatar on the generated page to the third multi-person virtual avatar corresponding to multiple user accounts includes: In response to an action posture update instruction triggered based on the updated first action description text, the first multi-person virtual avatar on the generated page will be updated to a third multi-person virtual avatar.
[0101] In one specific embodiment, the first action description text may include description text of the corresponding action postures of each virtual image in the first multi-person virtual image; optionally, human action posture features identified from the target person image can be converted into the first action description text through a large language model, etc.; specifically, the first update instruction may be an instruction to update the action description text corresponding to the third virtual image.
[0102] In one specific embodiment, updating the first multi-person virtual avatar on the generated page to the third multi-person virtual avatar corresponding to multiple user accounts includes: generating the third multi-person virtual avatar and updating the first multi-person virtual avatar on the generated page to the third multi-person virtual avatar corresponding to multiple user accounts.
[0103] In a specific embodiment, the detailed process of generating the third multi-person virtual avatar can be found in the detailed process of generating the first multi-person virtual avatar described above, and will not be repeated here. That is, the human body action posture features corresponding to the third virtual avatar are updated to the human body action posture features corresponding to the updated action description text of the third virtual avatar.
[0104] In the above embodiments, while the first multi-person virtual avatar is displayed on the generated page, the first action description text corresponding to each virtual avatar in the first multi-person virtual avatar is also displayed. By adjusting the action description text of the corresponding virtual avatar, the human body action posture features corresponding to the virtual avatar generation process can be updated, so as to realize the fine customization of the virtual avatar action posture in the multi-person virtual avatar generation scene, better meet the user's customization needs for the virtual avatar action posture, and thus better improve the realism and personalization of the virtual avatar's actions.
[0105] In an optional embodiment, the virtual avatar's pose can be updated by moving corresponding parts of the virtual avatar. Accordingly, the above-mentioned response to the action pose update instruction corresponding to the third virtual avatar in the first multi-person virtual avatar, updating the first multi-person virtual avatar in the generated page to the third multi-person virtual avatar corresponding to multiple user accounts may include: If a movement operation is detected targeting the target part of the third virtual avatar in the first multi-person virtual avatar, an action posture update command is triggered, and the first multi-person virtual avatar on the generated page is updated to the third multi-person virtual avatar.
[0106] In a specific embodiment, a preset state setting operation can be performed, such as clicking a preset state switching control, to update at least one preset part corresponding to each virtual image in the first multi-person virtual image from a different movement state to a movable state. Then, by dragging the target part in at least one preset part, the above-mentioned action posture update instruction can be triggered. Correspondingly, the third multi-person virtual image can be a multi-person virtual image after the target part of the third virtual image in the first multi-person virtual image has been moved.
[0107] In one specific embodiment, at least one preset part can be at least one pre-set movable part, such as at least one arm, leg, etc. The target part can be any of the at least one preset part.
[0108] In the above embodiments, after the first multi-person virtual avatar is generated, the action posture of the virtual avatar is updated by moving the target part of any virtual avatar. This enables fine-grained customization of the action posture of the virtual avatar in the multi-person virtual avatar generation scenario, better meets the user's customization needs for the action posture of the virtual avatar, and further enhances the realism and personalization of the virtual avatar's actions.
[0109] In one specific embodiment, when generating personalized multi-person virtual avatars (including a first multi-person virtual avatar, a second multi-person virtual avatar, and a third multi-person virtual avatar), works can be published based on the personalized multiple virtual avatars, or shared with other user accounts in multiple user accounts (i.e., user accounts other than the current account) through dialogue or other means.
[0110] In the above embodiments, after the first multi-person virtual avatar is generated, the action posture of the third virtual avatar in the first multi-person virtual avatar can be updated to realize the fine customization of the action posture of the virtual avatar in the multi-person virtual avatar generation scenario, better meet the user's customization needs for the action posture of the virtual avatar, and thus better improve the realism and personalization of the virtual avatar's action.
[0111] In an optional embodiment, before generating a first multi-user virtual avatar corresponding to multiple user accounts in response to a virtual avatar generation instruction triggered based on a target person image, the above method may further include: Displays descriptions of the second actions of multiple figures in the target image; In response to a second update instruction for the action description text corresponding to the target character in the second action description text, update the second action description text; Accordingly, the above-mentioned response to the virtual avatar generation instruction triggered based on the target person's image, generating a first multi-user virtual avatar corresponding to multiple user accounts includes: In response to the virtual avatar generation command, a first multi-person virtual avatar corresponding to the updated second action description text is generated. The action posture of each virtual avatar in the first multi-person virtual avatar corresponds one-to-one with the action posture of the corresponding user account described in the updated second action description text.
[0112] In one specific embodiment, the second action description text may include description text of the corresponding action posture of each person in the target person image; optionally, the human action posture features identified from the target person image can be converted into the second action description text using a large language model or the like; specifically, the second update instruction may be an instruction to update the action description text corresponding to the target person. The target person may be any one of multiple people in the target person image.
[0113] In one specific embodiment, the details of generating the first multi-person virtual avatar corresponding to the updated second action description text can be found in the above-mentioned descriptions, and will not be repeated here.
[0114] In the above embodiments, when the target person image is obtained, the second action description information of multiple people in the target person image is displayed. Then, by adjusting the action description text of the corresponding person, the corresponding human body action posture features in the virtual image generation process are updated, so as to realize the fine customization of the virtual image action posture in the multi-person virtual image generation scene, better meet the user's customization needs for the virtual image action posture, and further improve the realism and personalization of the virtual image's actions.
[0115] In one specific embodiment, after any personalized multi-person virtual avatar is generated, it can be saved and written to a shared generation record of multiple user accounts, so that any user account can choose the corresponding personalized multi-person virtual avatar for dressing up according to actual needs.
[0116] In an optional embodiment, the above method may further include: Display the first multi-person virtual avatar and at least one historical interaction information to present the interaction between corresponding user accounts based on at least some of the virtual avatars in the first multi-person virtual avatar; In one specific embodiment, at least one historical interaction information can be historical interaction information between at least some virtual avatars and their corresponding user accounts, such as the latest interaction information, frequently used interaction information, etc., and the interaction information type corresponding to at least one historical interaction information can include at least one interaction information type such as dialogue information, work interaction information, or live broadcast interaction information. Specifically, at least some virtual avatars can be some or all of the virtual avatars in a first group of virtual avatars. Correspondingly, at least one historical interaction information can be interaction information between some user accounts or interaction information between multiple user accounts.
[0117] In an optional embodiment, at least one piece of historical interaction information may be displayed together with the first multi-person virtual avatar on the generation page, or it may be displayed together with the first multi-person virtual avatar on other pages such as the virtual interactive space page.
[0118] In a specific embodiment, such as Figure 7 As shown, Figure 7 This is a schematic diagram of a page displaying a first set of virtual avatars of multiple people and at least one set of historical interaction information, according to an exemplary embodiment. The information corresponding to 701 can be multiple sets of historical interaction information.
[0119] In the above embodiments, when displaying the first multi-person virtual avatar, at least one historical interaction information is displayed to present the interaction between corresponding user accounts based on at least some of the virtual avatars in the first multi-person virtual avatar. This can combine the virtual avatars to more vividly simulate the historical interaction between user accounts, realize the linkage between historical interaction between users and multi-person virtual avatars, and better improve the interactivity between multiple user accounts.
[0120] In an optional embodiment, after generating a first multi-user virtual avatar corresponding to multiple user accounts in response to a virtual avatar generation instruction triggered based on a target person image, the above method may further include: In response to a dialogue interaction command triggered by the fourth virtual avatar, target dialogue information corresponding to the dialogue interaction command is sent to at least one other user account, so that when at least one other user account views the first multi-person virtual avatar, the target dialogue information is presented in the form of a dialogue between the fourth virtual avatar and the corresponding virtual avatar of at least one other user account. In one specific embodiment, the dialogue interaction instruction can be an instruction to interact (leave a message) with at least one other user account based on a fourth virtual avatar; optionally, the dialogue interaction instruction can be triggered by inputting target dialogue information based on the fourth virtual avatar and performing a confirmation operation to send it to at least one other user account, such as clicking the corresponding confirmation send control. Specifically, the fourth virtual avatar can be the virtual avatar corresponding to the current account in the first multi-person virtual avatar set; at least one other user account can be a user account other than the current account among multiple user accounts.
[0121] In one specific embodiment, for example, when any other user account views the first multi-person virtual avatar through the virtual interactive space page, the aforementioned target object information is displayed, such as... Figure 8 As shown, Figure 8 This is a schematic diagram illustrating a form in which target dialogue information is presented in the form of a fourth virtual avatar interacting with at least one other user account's corresponding virtual avatar, according to an exemplary embodiment; wherein, the virtual avatar corresponding to 801 can be the fourth virtual avatar in the first multi-person virtual avatar (the virtual avatar corresponding to the current account), the information corresponding to 802 can be the target dialogue information, and the virtual avatar corresponding to 803 can be another virtual avatar in the first multi-person virtual avatar (i.e., the virtual avatar corresponding to another user account).
[0122] In the above embodiments, in a multi-user virtual avatar interaction scenario shared by multiple users, a user's own virtual avatar can trigger the sending of dialogue information to at least one other user account. This dialogue information is presented in the form of a dialogue between the virtual avatar of the user account that sent the dialogue information and the virtual avatar of at least one other user account when the user account views the first multi-user virtual avatar. This can create a face-to-face dialogue atmosphere by combining virtual avatars, and better enhance the vividness of user message interaction.
[0123] In one optional embodiment, any one of the multiple user accounts can also customize the background style of pages displaying multiple virtual avatars, such as virtual avatar interaction pages and virtual space interaction pages; optionally, the background style can be automatically set to match the avatar style corresponding to the personalized customized multi-person virtual avatar itself.
[0124] Optionally, in this embodiment of the present disclosure, it can be used as an entry point for a touch-sensitive control or a non-touch-sensitive control. For example, it can be a non-touch-sensitive control that displays corresponding operation prompts. The operation prompts can be used to indicate the corresponding non-touch-sensitive operation. Taking the above-mentioned preset image customization entry point as an example, the operation prompts can be prompts for device control operations such as shaking to start customizing a multi-person virtual image, or prompts for corresponding voice control operations.
[0125] Optionally, in the scenario where another page is triggered to be displayed based on the display of one page in this embodiment of the present disclosure, the other page can be presented on a new page or displayed on the previous page in the form of a pop-up window or the like.
[0126] As can be seen from the technical solutions provided in the embodiments of this specification above, this specification can realize the personalized customization of multi-person virtual avatars corresponding to multiple accounts. In the process of personalized customization of multi-person virtual avatars, by obtaining character images including the characters corresponding to each of multiple user accounts, or the character images corresponding to each of multiple user accounts including multiple user accounts, the action postures of each virtual avatar in the multi-person virtual avatar can be personalized based on the action postures of the characters corresponding to each user account in the character images, restoring the real action postures of multiple users, better improving the realism and personalization of the action of the multi-person virtual avatar, and thus also better improving the interactivity based on the virtual avatar.
[0127] Figure 9 This is a block diagram illustrating a virtual avatar interaction device according to an exemplary embodiment. (Refer to...) Figure 9 The device includes: Page display module 910 is configured to display a virtual avatar interaction page shared by multiple user accounts. The virtual avatar interaction page includes a preset avatar customization entry. The character image acquisition module 920 is configured to execute a character image acquisition instruction triggered based on a preset image customization entry, and acquire a target character image. The target character image is either a character image including the corresponding character of multiple user accounts, or a character image corresponding to multiple user accounts. The first image generation module 930 is configured to execute a virtual image generation instruction triggered based on the target person image to generate a first multi-person virtual image corresponding to multiple user accounts, and the action posture of each virtual image in the first multi-person virtual image corresponds one-to-one with the action posture of the corresponding user account in the target person image.
[0128] In an optional embodiment, the first multi-person virtual avatar is generated by a preset generation model based on the target person image and the human body action and posture features corresponding to the user accounts in the target person image. The human body action and posture features are obtained by human body posture recognition of the target person image.
[0129] In an optional embodiment, the preset generation model includes an appearance feature extraction layer, multiple appearance action learning layers, and a virtual avatar generation layer; the appearance feature extraction layer is used to extract the appearance features corresponding to each user account from the target person image; each appearance action learning layer is used to learn the appearance action of a user account based on the appearance features of one user account and the human body action posture features of one user account, to obtain the appearance action features of a user account; the virtual avatar generation layer is used to generate a first multi-person virtual avatar based on the appearance action features of the user accounts.
[0130] In an optional embodiment, the above-described apparatus further includes: The style information display module is configured to display at least one style option and / or a custom style control; The style determination module is configured to execute a style setting instruction triggered in response to at least one style option or a custom style control to determine the target image style. The first image generation module 930 includes: The first image generation unit is configured to execute a virtual image generation instruction to generate a first multi-person virtual image corresponding to the target image style.
[0131] In an optional embodiment, the custom style control is a custom style input control; the style information display module is further configured to display at least one style option and / or include the custom style input control, and the custom style input control includes reference style description information; Among them, the reference style description information is style description information extracted from the target person image.
[0132] In an optional embodiment, the above-described apparatus further includes: The prompt generation module is configured to display the prompt information for the first multi-person virtual avatar and the viewing entry for the first multi-person virtual avatar on the dialogue pages corresponding to multiple user accounts.
[0133] In an optional embodiment, the first image generation module 930 includes: The second image generation unit is configured to execute a virtual image generation instruction in response to the virtual image generation instruction, and generate a first multi-person virtual image when the human body action posture features corresponding to multiple user accounts meet the preset action constraint conditions. Among them, the preset action constraints include action confidence constraints and / or action specification constraints.
[0134] In an optional embodiment, the above-described apparatus further includes: The second image generation module is configured to execute a response to the virtual image generation command, and generate a second multi-person virtual image corresponding to multiple user accounts when the human body action posture features of any user account among multiple user accounts do not meet the preset action constraints. Among them, the action posture of the first virtual image in the second multi-person virtual image is matched one-to-one with the action posture of the corresponding user account in the target person image, and the action posture of the second virtual image in the second multi-person virtual image is the corresponding preset action posture; the first virtual image is the virtual image whose human body action posture features of the corresponding user account in the second multi-person virtual image meet the preset action constraint conditions; the second virtual image is the virtual image whose human body action posture features of the corresponding user account in the second multi-person virtual image do not meet the preset action constraint conditions.
[0135] In an optional embodiment, the first image generation module 930 includes: The virtual avatar display module is configured to respond to the virtual avatar generation command and display the first multi-person virtual avatar on the generation page; The above-mentioned device also includes: The virtual avatar update unit is configured to execute an action posture update instruction in response to the third virtual avatar in the first multi-person virtual avatar, and update the first multi-person virtual avatar on the generated page to the third multi-person virtual avatar corresponding to multiple user accounts. Among them, the third virtual image is the virtual image whose action posture is updated by the action posture update command in the first multi-person virtual image; the third multi-person virtual image is the multi-person virtual image after updating the action posture corresponding to the third virtual image in the first multi-person virtual image to the action posture indicated by the action posture update command.
[0136] In an optional embodiment, the virtual avatar display module includes: The first information display unit is configured to display the first multi-person virtual avatar and the first action description text corresponding to each virtual avatar in the first multi-person virtual avatar on the generated page; The action update unit is configured to execute a first update instruction in response to the action description text corresponding to the third virtual image in the first action description text, and update the first action description text. The virtual avatar update unit includes: The first update unit is configured to execute an action posture update instruction triggered based on the updated first action description text, and update the first multi-person virtual avatar in the generated page to the third multi-person virtual avatar.
[0137] In an optional embodiment, the virtual avatar update unit includes: The second update unit is configured to, upon detecting a movement operation on the target part corresponding to the third virtual avatar in the first multi-person virtual avatar, trigger an action posture update instruction to update the first multi-person virtual avatar in the generated page to the third multi-person virtual avatar.
[0138] In an optional embodiment, the above-described apparatus further includes: The description text display module is configured to perform second action description information for multiple figures in the target figure image; The action update module is configured to execute a second update instruction in response to the action description text corresponding to the target character in the second action description text, and update the second action description text. The first image generation module 930 includes: The third image generation unit is configured to execute a response to the virtual image generation instruction, generating a first multi-person virtual image corresponding to the updated second action description text. The action posture of each virtual image in the first multi-person virtual image corresponds one-to-one with the character action posture of the corresponding user account described in the updated second action description text.
[0139] In an optional embodiment, the above-described apparatus further includes: The historical interaction module is configured to execute a response to the virtual avatar generation instruction, displaying a first multi-person virtual avatar and at least one historical interaction information, so as to present the interaction between corresponding user accounts based on at least some of the virtual avatars in the first multi-person virtual avatar. Among them, at least one piece of historical interaction information is historical interaction information between at least some of the user accounts corresponding to the virtual avatars.
[0140] In an optional embodiment, the above-described apparatus further includes: The dialogue interaction module is configured to execute a dialogue interaction command triggered based on the fourth virtual avatar, and send the target dialogue information corresponding to the dialogue interaction command to at least one other user account, so that when at least one other user account views the first multi-person virtual avatar, the target dialogue information is presented in the form of a dialogue between the fourth virtual avatar and the corresponding virtual avatar of at least one other user account. Among them, the fourth virtual avatar is the virtual avatar corresponding to the current account in the first multi-person virtual avatar; at least one other user account is a user account other than the current account among multiple user accounts.
[0141] In an optional embodiment, the above-described apparatus further includes: The avatar customization prompt module is configured to display the virtual avatar interaction entry and customization unlock prompt information on the dialogue page corresponding to multiple user accounts when the interaction information between multiple user accounts meets the preset interaction conditions. The customization unlock prompt information is used to indicate that the virtual avatar customization function has been successfully unlocked. The page display module includes: The page display unit is configured to execute a page display instruction triggered by the virtual avatar interaction entry, and display a virtual avatar interaction page including a preset avatar customization entry.
[0142] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.
[0143] Figure 10 This is a block diagram illustrating an electronic device for interacting with a virtual avatar, according to an exemplary embodiment. The electronic device may be a terminal, and its internal structure diagram may be as follows: Figure 10 As shown. This electronic device can be a terminal, and its internal structure diagram can be as follows. Figure 10 As shown. This electronic device may include an RF (Radio Frequency) circuit 1010, a memory 1020 including one or more computer-readable storage media, an input unit 1030, a display unit 1040, a sensor 1050, an audio circuit 1060, a WiFi (Wireless Fidelity) module 1070, a processor 1080 including one or more processing cores, and a power supply 1090, among other components. Those skilled in the art will understand that... Figure 10 The terminal structure shown does not constitute a limitation on the terminal and may include more or fewer components than shown, or combine certain components, or have different component arrangements. Wherein: The RF circuit 1010 can be used for receiving and transmitting signals during information transmission or calls. Specifically, it receives downlink information from the base station and hands it over to one or more processors 1080 for processing; additionally, it transmits uplink data to the base station. Typically, the RF circuit 1010 includes, but is not limited to, an antenna, at least one amplifier, a tuner, one or more oscillators, a Subscriber Identity Module (SIM) card, a transceiver, a coupler, an LNA (Low Noise Amplifier), a duplexer, etc. Furthermore, the RF circuit 1010 can also communicate wirelessly with networks and other terminals. Wireless communication can use any communication standard or protocol, including but not limited to GSM (Global System for Mobile communication), GPRS (General Packet Radio Service), CDMA (Code Division Multiple Access), WCDMA (Wideband Code Division Multiple Access), LTE (Long Term Evolution), email, SMS (Short Messaging Service), etc.
[0144] The memory 1020 can be used to store software programs and modules. The processor 1080 executes various functional applications and data processing by running the software programs and modules stored in the memory 1020. The memory 1020 may mainly include a program storage area and a data storage area. The program storage area may store the operating system, application programs required for the functions, etc.; the data storage area may store data created according to the use of the terminal, etc. In addition, the memory 1020 may include high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device. Accordingly, the memory 1020 may also include a memory controller to provide access to the memory 1020 for the processor 1080 and the input unit 1030.
[0145] The input unit 1030 can be used to receive input numerical or character information, and to generate keyboard, mouse, joystick, optical, or trackball signal inputs related to user settings and function control. Specifically, the input unit 1030 may include a touch-sensitive surface 1031 and other input devices 1032. The touch-sensitive surface 1031, also known as a touch display screen or touchpad, can collect touch operations performed by the user on or near it (such as operations performed by the user using a finger, stylus, or any suitable object or nearby on or near the touch-sensitive surface 1031), and drive the corresponding connection device according to a pre-set program. Optionally, the touch-sensitive surface 1031 may include two parts: a touch detection device and a touch controller. The touch detection device detects the user's touch position and the signal generated by the touch operation, and transmits the signal to the touch controller; the touch controller receives touch information from the touch detection device, converts it into touch point coordinates, sends it to the processor 1080, and can receive and execute commands from the processor 1080. In addition, the touch-sensitive surface 1031 can be implemented using various types such as resistive, capacitive, infrared, and surface acoustic wave. Besides the touch-sensitive surface 1031, the input unit 1030 may also include other input devices 1032. Specifically, other input devices 1032 may include, but are not limited to, one or more of the following: a physical keyboard, function keys (such as volume control buttons, power buttons, etc.), a trackball, a mouse, and a joystick. The display unit 1040 can be used to display information input by the user or information provided to the user, as well as various graphical user interfaces of the terminal. These graphical user interfaces can be composed of graphics, text, icons, video, and any combination thereof. The display unit 1040 may include a display panel 1041, which may optionally be configured as an LCD (Liquid Crystal Display), OLED (Organic Light-Emitting Diode), or similar display panel. Further, a touch-sensitive surface 1031 may cover the display panel 1041. When the touch-sensitive surface 1031 detects a touch operation on or near it, it transmits the information to the processor 1080 to determine the type of touch event. Subsequently, the processor 1080 provides corresponding visual output on the display panel 1041 according to the type of touch event. The touch-sensitive surface 1031 and the display panel 1041 can be two independent components to achieve input and output functions. However, in some embodiments, the touch-sensitive surface 1031 and the display panel 1041 can be integrated to achieve input and output functions.
[0146] The terminal may also include at least one sensor 1050, such as a light sensor, a motion sensor, and other sensors. Specifically, the light sensor may include an ambient light sensor and a proximity sensor. The ambient light sensor can adjust the brightness of the display panel 1041 according to the ambient light level, and the proximity sensor can turn off the display panel 1041 and / or the backlight when the terminal is moved to the ear. As a type of motion sensor, a gravity acceleration sensor can detect the magnitude of acceleration in various directions (generally three axes). When stationary, it can detect the magnitude and direction of gravity and can be used for applications that identify the terminal's posture (such as landscape / portrait switching, related games, magnetometer posture calibration), vibration recognition-related functions (such as pedometer, tapping), etc. Other sensors that may be configured on the terminal, such as gyroscopes, barometers, hygrometers, thermometers, and infrared sensors, will not be described in detail here.
[0147] Audio circuitry 1060, speaker 1061, and microphone 1062 provide an audio interface between the user and the terminal. Audio circuitry 1060 converts received audio data into electrical signals, which are then transmitted to speaker 1061, where they are converted into sound signals for output. Conversely, microphone 1062 converts collected sound signals into electrical signals, which are received by audio circuitry 1060, converted back into audio data, and then output to processor 1080 for processing. The processed audio data is then transmitted via RF circuitry 1010 to, for example, another terminal, or output to memory 1020 for further processing. Audio circuitry 1060 may also include an earphone jack to facilitate communication between a peripheral headset and the terminal.
[0148] WiFi is a short-range wireless transmission technology. This terminal, through the WiFi module 1070, can help users send and receive emails, browse web pages, and access streaming media, providing users with wireless broadband internet access. Although Figure 10 The WiFi module 1070 is shown, but it is understood that it is not a necessary component of the terminal and can be omitted as needed without changing the nature of the invention.
[0149] The processor 1080 is the control center of the terminal, connecting various parts of the terminal through various interfaces and lines. It executes software programs and / or modules stored in the memory 1020 and calls data stored in the memory 1020 to perform various functions and process data, thereby enabling overall monitoring of the terminal. Optionally, the processor 1080 may include one or more processing cores; preferably, the processor 1080 may integrate an application processor and a modem processor, wherein the application processor mainly handles the operating system, user page, and applications, while the modem processor mainly handles wireless communication. It is understood that the modem processor may not be integrated into the processor 1080.
[0150] The terminal also includes a power supply 1090 (such as a battery) to power various components. Preferably, the power supply can be logically connected to the processor 1080 through a power management system, thereby enabling functions such as charging, discharging, and power consumption management through the power management system. The power supply 1090 may also include one or more DC or AC power supplies, recharging systems, power fault detection circuits, power converters or inverters, power status indicators, and other arbitrary components.
[0151] Although not shown, the terminal may also include a camera, Bluetooth module, etc., which will not be described in detail here. Specifically, in this embodiment, the display unit of the terminal is a touch screen display, and the terminal also includes a memory and one or more programs, wherein one or more programs are stored in the memory and configured to be executed by one or more processors of the instructions in the method embodiment of the present invention.
[0152] In an exemplary embodiment, an electronic device is also provided, including: a processor; and a memory for storing processor-executable instructions; wherein the processor is configured to execute the instructions to implement the virtual avatar interaction method as described in the embodiments of this disclosure.
[0153] In an exemplary embodiment, a computer-readable storage medium is also provided, wherein when the instructions in the storage medium are executed by a processor of an electronic device, the electronic device is enabled to perform the virtual avatar interaction method of the present disclosure embodiments. In an exemplary embodiment, a computer program product comprising computer program instructions is also provided, which, when executed by a processor on a computer, causes the computer to perform the virtual avatar interaction method of the present disclosure embodiments.
[0154] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this disclosure can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM), etc.
[0155] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This disclosure is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the following claims.
[0156] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.
Claims
1. A method for interactive use of virtual avatars, characterized in that, include: Displays a virtual avatar interaction page shared by multiple user accounts, the virtual avatar interaction page including a preset avatar customization entry; In response to a character image acquisition instruction triggered based on the preset image customization entry, a target character image is acquired. The target character image is either a character image including the corresponding character of each of the multiple user accounts, or a character image corresponding to each of the multiple user accounts. In response to a virtual avatar generation instruction triggered based on the target person image, a first multi-person virtual avatar corresponding to the multiple user accounts is generated, and the action posture of each virtual avatar in the first multi-person virtual avatar corresponds one-to-one with the action posture of the corresponding user account in the target person image.
2. The virtual avatar interaction method according to claim 1, characterized in that, The first multi-person virtual avatar is generated by a preset generation model based on the target person image and the human body action and posture features corresponding to the user accounts in the target person image. The human body action and posture features are obtained by human body posture recognition of the target person image.
3. The virtual avatar interaction method according to claim 2, characterized in that, The preset generation model includes an appearance feature extraction layer, multiple appearance action learning layers, and a virtual avatar generation layer. The appearance feature extraction layer is used to extract the appearance features corresponding to each of the multiple user accounts from the target person image. Each appearance action learning layer is used to learn the appearance action of a person based on the appearance feature of one user account and the human body action posture feature of the user account, to obtain the appearance action feature of the user account. The virtual avatar generation layer is used to generate the first multi-person virtual avatar based on the appearance action features of the multiple user accounts.
4. The virtual avatar interaction method according to any one of claims 1 to 3, characterized in that, Before generating the first multi-user virtual avatar corresponding to the multiple user accounts in response to the virtual avatar generation instruction triggered based on the target person image, the method further includes: Display at least one style option and / or a custom style control; In response to a style setting instruction triggered based on any of the at least one style option or the custom style control, the target image style is determined; The step of generating a first multi-user virtual avatar corresponding to the multiple user accounts in response to a virtual avatar generation instruction triggered based on the target person image includes: In response to the virtual avatar generation instruction, the first multi-person virtual avatar corresponding to the target avatar style is generated.
5. The virtual avatar interaction method according to claim 4, characterized in that, The custom style control is a custom style input control; The provision of at least one style option and / or custom style control includes: Display at least one style option and / or include the custom style input control, wherein the custom style input control includes reference style description information; The reference style description information is style description information extracted based on the target person image.
6. The virtual avatar interaction method according to any one of claims 1 to 3, characterized in that, After generating the first multi-user virtual avatars corresponding to the multiple user accounts in response to the virtual avatar generation instruction triggered based on the target person image, the method further includes: The dialog pages corresponding to the multiple user accounts display the generation prompt information for the first multi-person virtual avatar and the viewing entry for the first multi-person virtual avatar.
7. The virtual avatar interaction method according to claim 2, characterized in that, The step of generating a first multi-user virtual avatar corresponding to the multiple user accounts in response to a virtual avatar generation instruction triggered based on the target person image includes: In response to the virtual avatar generation instruction, the first multi-person virtual avatar is generated when the human body action posture features corresponding to the multiple user accounts all meet the preset action constraint conditions. The preset action constraints include action confidence constraints and / or action specification constraints.
8. The virtual avatar interaction method according to claim 7, characterized in that, The method further includes: In response to the virtual avatar generation instruction, if the human body action posture features corresponding to any of the multiple user accounts do not meet the preset action constraint conditions, a second multi-person virtual avatar corresponding to the multiple user accounts is generated. In this context, the action posture of the first virtual image in the second multi-person virtual avatar corresponds one-to-one with the action posture of the corresponding user account in the target character image; the action posture of the second virtual image in the second multi-person virtual avatar is the corresponding preset action posture; the first virtual image is a virtual image whose human body action posture features of the corresponding user account in the second multi-person virtual avatar satisfy the preset action constraint conditions; and the second virtual image is a virtual image whose human body action posture features of the corresponding user account in the second multi-person virtual avatar do not satisfy the preset action constraint conditions.
9. The virtual avatar interaction method according to any one of claims 1 to 3, characterized in that, The step of generating a first multi-user virtual avatar corresponding to the multiple user accounts in response to a virtual avatar generation instruction triggered based on the target person image includes: In response to the virtual avatar generation instruction, the first multi-person virtual avatar is displayed on the generation page; The method further includes: In response to the action posture update instruction corresponding to the third virtual image in the first multi-person virtual image, the first multi-person virtual image in the generated page is updated to the third multi-person virtual image corresponding to the multiple user accounts; The third virtual image is the virtual image corresponding to the action posture updated by the action posture update instruction in the first multi-person virtual image; the third multi-person virtual image is the multi-person virtual image after updating the action posture corresponding to the third virtual image in the first multi-person virtual image to the action posture indicated by the action posture update instruction.
10. The virtual avatar interaction method according to claim 9, characterized in that, The step of displaying the first multi-person virtual avatar on the generation page in response to the virtual avatar generation instruction includes: The generated page displays the first multi-person virtual avatar and the first action description text corresponding to each virtual avatar in the first multi-person virtual avatar; In response to a first update instruction for the action description text corresponding to the third virtual character in the first action description text, the first action description text is updated; The step of updating the first multi-person virtual avatar in the generated page to the third multi-person virtual avatar corresponding to the multiple user accounts in response to the action posture update instruction for the third multi-person virtual avatar in the first multi-person virtual avatar includes: In response to the action posture update instruction triggered based on the updated first action description text, the first multi-person virtual avatar in the generated page is updated to the third multi-person virtual avatar.
11. The virtual avatar interaction method according to claim 9, characterized in that, The step of updating the first multi-person virtual avatar in the generated page to the third multi-person virtual avatar corresponding to the multiple user accounts in response to the action posture update instruction for the third multi-person virtual avatar in the first multi-person virtual avatar includes: If a movement operation is detected targeting the target part of the third virtual avatar in the first multi-person virtual avatar, the action posture update instruction is triggered to update the first multi-person virtual avatar in the generated page to the third multi-person virtual avatar.
12. The virtual avatar interaction method according to any one of claims 1 to 3, characterized in that, Before generating the first multi-user virtual avatar corresponding to the multiple user accounts in response to the virtual avatar generation instruction triggered based on the target person image, the method further includes: Display the second action description information of the multiple figures in the target figure image; In response to a second update instruction for the action description text corresponding to the target character in the second action description text, update the second action description text; The step of generating a first multi-user virtual avatar corresponding to the multiple user accounts in response to a virtual avatar generation instruction triggered based on the target person image includes: In response to the virtual avatar generation instruction, the first multi-person virtual avatar corresponding to the updated second action description text is generated, and the action posture of each virtual avatar in the first multi-person virtual avatar corresponds one-to-one with the action posture of the corresponding user account described in the updated second action description text.
13. The virtual avatar interaction method according to any one of claims 1 to 3, characterized in that, The method further includes: Display the first multi-person virtual avatar and at least one historical interaction information to present the interaction between corresponding user accounts based on at least some of the virtual avatars in the first multi-person virtual avatar; Wherein, the at least one historical interaction information is the historical interaction information between the user accounts corresponding to the at least some virtual images.
14. The virtual avatar interaction method according to any one of claims 1 to 3, characterized in that, After generating the first multi-user virtual avatars corresponding to the multiple user accounts in response to the virtual avatar generation instruction triggered based on the target person image, the method further includes: In response to a dialogue interaction command triggered by a fourth virtual avatar, target dialogue information corresponding to the dialogue interaction command is sent to at least one other user account, so that when the at least one other user account views the first multi-person virtual avatar, the target dialogue information is presented in the form of a dialogue between the fourth virtual avatar and the virtual avatar corresponding to the at least one other user account. The fourth virtual avatar is the virtual avatar corresponding to the current account among the first multiple virtual avatars; the at least one other user account is a user account other than the current account among the multiple user accounts.
15. The virtual avatar interaction method according to any one of claims 1 to 3, characterized in that, The method further includes: When the interaction information between the multiple user accounts meets the preset interaction conditions, the virtual avatar interaction entry and customized unlock prompt information are displayed on the dialogue page corresponding to the multiple user accounts. The customized unlock prompt information is used to indicate that the virtual avatar customization function has been successfully unlocked. The interactive page displaying the shared virtual avatars of multiple user accounts includes: In response to a page display instruction triggered by the virtual avatar interaction entry, the virtual avatar interaction page, including the preset avatar customization entry, is displayed.
16. A virtual avatar interactive device, characterized in that, include: The page display module is configured to execute and display a virtual avatar interaction page shared by multiple user accounts, the virtual avatar interaction page including a preset avatar customization entry; The character image acquisition module is configured to execute a character image acquisition instruction triggered based on the preset image customization entry, and acquire a target character image, wherein the target character image is a character image including the character corresponding to each of the multiple user accounts, or is a character image corresponding to each of the multiple user accounts. The first image generation module is configured to execute a virtual image generation instruction triggered based on the target person image to generate a first multi-person virtual image corresponding to the multiple user accounts, and the action posture of each virtual image in the first multi-person virtual image corresponds one-to-one with the action posture of the corresponding user account in the target person image.
17. An electronic device, characterized in that, include: processor; Memory used to store the processor's executable instructions; The processor is configured to execute the instructions to implement the virtual avatar interaction method as described in any one of claims 1 to 15.
18. A computer-readable storage medium, characterized in that, When the instructions in the storage medium are executed by the processor of the electronic device, the electronic device is able to perform the virtual avatar interaction method as described in any one of claims 1 to 15.
19. A computer program product, characterized in that, It includes computer program instructions that, when executed by a computer's processor, cause the computer to perform the virtual avatar interaction method as described in any one of claims 1 to 15.
