Virtual image generation method and device based on AR scene and storage medium

Abstract

The invention discloses a virtual image generation method and device based on an AR scene and a storage medium, and relates to the field of meta universe. The method comprises the steps that image style preference information of a target user and a virtual image label of the target user are acquired, and the virtual image label describes image features of a corresponding virtual image for the target user; determining N alternative materials from a virtual image material library according to the image style preference information and the virtual image label, N being an integer greater than 1; m virtual images are generated according to the N alternative materials, M is an integer larger than or equal to 1, and each virtual image is composed of at least one alternative material in the N alternative materials; and adjusting the virtual image according to the scene information of the AR scene to obtain a target virtual image corresponding to the target user. The technical problem that a personalized virtual image in an AR scene cannot be generated in the prior art is solved.

Description

Technical Field

[0001] The present application relates to the field of metaverse, and more specifically, to a method, device and storage medium for generating a virtual image based on an AR scene. Background Art

[0002] In today's digital age, augmented reality (AR) technology is rapidly changing users' lifestyles, entertainment methods, and ways of understanding the world with its unique immersive experience and reality enhancement capabilities. AR technology is not limited to the field of games, it also shows great potential in many fields such as education, medical care, tourism, retail, and social interaction. With the continuous development of AR technology, users' demand for personalized virtual experience is growing, especially in the creation of virtual images. Users hope to have a virtual avatar that is closer to their own style, emotional expression, and cultural background to enhance their immersion and interactive experience in AR scenes.

[0003] However, existing technologies still have many limitations in generating personalized virtual images: reliance on fixed templates or preset libraries, low user participation, poor integration, failure to fully consider the real-time dynamic information of the user's environment, etc., so it is impossible to generate suitable personalized virtual images in AR scenes.

[0004] To address the above-mentioned problems, no effective solution has been proposed yet. Summary of the invention

[0005] The main purpose of the present application is to provide a method, device and storage medium for generating a virtual image based on an AR scene, so as to at least solve the technical problem in the prior art that a personalized virtual image in an AR scene cannot be generated.

[0006] In order to achieve the above-mentioned purpose, according to one aspect of an embodiment of the present application, a method for generating a virtual image based on an AR scene is provided, including: obtaining image style preference information of a target user and a virtual image tag of the target user, wherein the virtual image tag is an image feature of the virtual image corresponding to the target user; determining N alternative materials from a virtual image material library according to the image style preference information and the virtual image tag, wherein N is an integer greater than 1; generating M virtual images according to the N alternative materials, wherein M is an integer greater than or equal to 1, and the virtual image is composed of at least one alternative material among the N alternative materials; adjusting the virtual image according to the scene information of the AR scene to obtain a target virtual image corresponding to the target user.

[0007] Optionally, obtaining the target user's image style preference information includes: obtaining the target user's historical behavior information in the AR scene, wherein the historical behavior information at least includes a virtual image selected by the target user in the AR scene, information about activities the target user has participated in in the AR scene, and social interaction information about the virtual image posted by the target user in the AR scene; and determining the target user's image style preference information from the historical behavior information.

[0008] Optionally, obtaining the virtual image tag of the target user includes: obtaining a target text and a target image submitted by the target user, wherein the target text records the target user's description information about the desired virtual image, and the target image includes the virtual image that the target user expects to use for reference; and extracting the virtual image tag of the target user from the target text and the target image.

[0009] Optionally, based on the image style preference information and the virtual image label, N candidate materials are determined from the virtual image material library, including: generating a preference feature label of the target user based on the image style preference information; fusing the preference feature label and the virtual image label into an overall image style label of the target user, and determining N candidate materials from the virtual image material library based on the overall image style label, wherein the overall image style label includes at least one of the following dimensions: overall image style information, local style information, painting style information, and color information.

[0010] Optionally, determining N candidate materials from a virtual image material library according to the overall image style label includes: obtaining a matching score of each virtual image material in the virtual image material library in each dimension of the overall image style label, wherein the virtual image material library includes pre-classified virtual image materials of multiple styles; performing feature matching with the overall image style label in each dimension for materials of the same type according to the matching score, and screening out X groups of virtual image materials in each dimension that meet the overall image style label, wherein X is an integer greater than 1; and taking an intersection of the screened X groups of virtual image materials to obtain a candidate material set consisting of the N candidate materials.

[0011] Optionally, the virtual image is adjusted according to scene information of the AR scene, including: obtaining real environment information of the target user through the camera and sensor of the AR device, and adjusting the proportion and position of the virtual image according to the real environment information of the target user; and / or, obtaining lighting conditions in the AR scene, and adjusting the material properties and lighting position of the virtual image according to the lighting conditions.

[0012] Optionally, the image style preference information includes at least one of the following style preference information: clothing style preference information of the virtual image; physical appearance style preference information of the virtual image; background environment style preference information of the virtual image; and posture style preference information of the virtual image.

[0013] In order to achieve the above-mentioned purpose, according to another aspect of an embodiment of the present application, a virtual image generation device based on an AR scene is also provided, including: an acquisition unit, which acquires the image style preference information of the target user and the virtual image label of the target user, wherein the virtual image label is the image feature of the virtual image corresponding to the target user; a determination unit, which determines N alternative materials from a virtual image material library according to the image style preference information and the virtual image label, wherein N is an integer greater than 1; a generation unit, which generates M virtual images according to the N alternative materials, wherein M is an integer greater than or equal to 1, and the virtual image is composed of at least one alternative material among the N alternative materials; an adjustment unit, which adjusts the virtual image according to the scene information of the AR scene to obtain a target virtual image corresponding to the target user.

[0014] Optionally, the acquisition unit includes: a first acquisition subunit and a first determination subunit. The first acquisition subunit is used to acquire the historical behavior information of the target user in the AR scene, wherein the historical behavior information at least includes the virtual image selected by the target user in the AR scene, the activity information participated by the target user in the AR scene, and the social interaction information about the virtual image published by the target user in the AR scene; the first determination subunit is used to determine the image style preference information of the target user from the historical behavior information.

[0015] Optionally, the acquisition unit includes: a second acquisition subunit and a first extraction subunit. The second acquisition subunit is used to acquire a target text and a target image submitted by a target user, wherein the target text records the description information of the target user on the desired virtual image, and the target image includes the virtual image that the target user expects to use for reference; and the first extraction subunit is used to extract the virtual image label of the target user from the target text and the target image.

[0016] Optionally, the determination unit includes: a first generation subunit and a second determination subunit. The first generation subunit is used to generate a preference feature label of the target user according to the image style preference information; the second determination subunit is used to merge the preference feature label and the virtual image label into an overall image style label of the target user, and determine N candidate materials from the virtual image material library according to the overall image style label, wherein the overall image style label includes at least one of the following dimensions: overall image style information, local style information, painting style information, and color information.

[0017] Optionally, the second determination subunit includes: a first acquisition module, a first processing module, and a second processing module. The first acquisition module is used to obtain the matching score of each virtual image material in the virtual image material library in each dimension of the overall image style label, wherein the virtual image material library includes virtual image materials of multiple styles pre-classified; the first processing module is used to perform feature matching with the overall image style label in each dimension for the same type of material according to the matching score, and screen out X groups of virtual image materials in each dimension that meet the overall image style label, wherein X is an integer greater than 1; the second processing module is used to take the intersection of the screened X groups of virtual image materials to obtain a candidate material set consisting of N candidate materials.

[0018] Optionally, the adjustment unit includes: a third acquisition subunit and a fourth acquisition subunit. The third acquisition subunit is used to acquire the real environment information of the target user through the camera and sensor of the AR device, and adjust the proportion and position of the virtual image according to the real environment information of the target user; the fourth acquisition subunit is used to acquire the lighting conditions in the AR scene, and adjust the material properties and lighting position of the virtual image according to the lighting conditions.

[0019] Optionally, the image style preference information includes at least one of the following style preference information: clothing style preference information of the virtual image; physical appearance style preference information of the virtual image; background environment style preference information of the virtual image; and posture style preference information of the virtual image.

[0020] According to another aspect of an embodiment of the present application, an electronic device is also provided, including: a memory storing an executable program; and a processor for running the program, wherein the above-mentioned method for generating a virtual image based on an AR scene is executed when the program is running.

[0021] According to another aspect of an embodiment of the present application, a computer-readable storage medium is also provided, the computer-readable storage medium including a stored executable program, wherein when the executable program is running, the device where the computer-readable storage medium is located is controlled to execute the above-mentioned AR scene-based virtual image generation method.

[0022] According to another aspect of an embodiment of the present application, a computer program product is also provided, including computer instructions, which, when executed by a processor, implement the steps of the above-mentioned method for generating a virtual image based on an AR scene.

[0023] In an embodiment of the present application, firstly, the image style preference information of the target user and the virtual image label of the target user are obtained, wherein the virtual image label is the image feature of the virtual image corresponding to the target user's description, and then, based on the image style preference information and the virtual image label, N candidate materials are determined from the virtual image material library, wherein N is an integer greater than 1, and then, based on the N candidate materials, M virtual images are generated, wherein M is an integer greater than or equal to 1, and the virtual image is composed of at least one candidate material among the N candidate materials, and finally, the virtual image is adjusted according to the scene information of the AR scene to obtain the target virtual image corresponding to the target user.

[0024] From the above content, it can be seen that this application comprehensively quantifies the user's style preferences by analyzing the user's historical behavior data in the AR scene, such as the virtual image selected by the user, the activities participated in, and the interactive information in the community. In addition, the user can actively describe the characteristics of the virtual image he or she expects and set labels for the virtual image.

[0025] Secondly, based on the style preference information and virtual image labels obtained in the previous step, N alternative materials are intelligently selected from the pre-classified virtual image material library, where N is an integer greater than 1. These materials cover multiple dimensions of user preferences, such as overall image style, local style, painting style and color information.

[0026] In addition, M different virtual images are generated using the selected N candidate materials, where M is an integer greater than or equal to 1. Each generated virtual image is composed of at least one candidate material, ensuring that the generated virtual image can fully reflect the personalized needs of the user.

[0027] Finally, the generated virtual image is adjusted to adapt to the scene based on the real-time information of the target user's AR scene, such as the real environment, lighting conditions, etc. This includes not only adjusting the scale and position of the virtual image to match the user's real environment, but also adjusting the material properties and lighting effects to enhance the visual integration of the virtual image and the AR scene.

[0028] It can be seen that through the technical solution of the present application, the purpose of achieving a deep understanding of user style preferences is achieved, thereby achieving the technical effect of ensuring that the generated virtual image meets personalized needs and enhancing the user's immersive experience and interactive fun in the AR environment, thereby solving the technical problem that the existing technology cannot generate personalized virtual images in AR scenes. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0030] Figure 1 A hardware structure block diagram of a computer terminal for implementing a method for generating a virtual image based on an AR scene is shown;

[0031] Figure 2 is a flowchart of an optional method for generating a virtual image based on an AR scene according to an embodiment of the present application;

[0032] Figure 3 is a schematic diagram of a virtual image generation device based on an AR scene provided according to an embodiment of the present application;

[0033] Figure 4 It is a structural block diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0034] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0035] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0036] It should also be noted that the collected information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for display, data for analysis, etc.) involved in this application are information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of relevant data are in compliance with relevant laws, regulations and standards, necessary confidentiality measures are taken, and public order and good customs are not violated, and corresponding operation entrances are provided for users to choose to authorize or refuse. For example, an interface is set up between this system and relevant users or institutions to provide users with corresponding operation entrances for users to choose to agree or refuse the results of automated decision-making; if the user chooses to refuse, the expert decision-making process will be entered.

[0037] According to an embodiment of the present application, a method embodiment of a method for generating a virtual image based on an AR scene is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0038] It should be noted that a virtual image generation system can be used as the execution subject of the virtual image generation method based on AR scene in the embodiment of the present application. It is understandable that the virtual image generation method based on AR scene provided in the embodiment of the present application can also be executed by other systems or devices, and the embodiment of the present application does not specifically limit this.

[0039] The method embodiments provided in the embodiments of the present application can be executed in a mobile terminal, a computer terminal or a similar computing device. Figure 1 The hardware structure block diagram of a computer terminal (or mobile device) for implementing a method for generating a virtual image based on an AR scene is shown. Figure 1 As shown, the computer terminal 10 (or mobile device) may include one or more (102a, 102b, ..., 102n are used to illustrate) processors 102 (the processor 102 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA), a memory 104 for storing data, and a transmission device 106 for communication functions. In addition, it may also include: a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of the BUS bus), a network interface, a power supply and / or a camera. It can be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the above electronic device. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations shown.

[0040] It should be noted that the above-mentioned one or more processors 102 and / or other data processing circuits may generally be referred to as "data processing circuits" in this article. The data processing circuit may be embodied in whole or in part as software, hardware, firmware or any other combination. In addition, the data processing circuit may be a single independent processing module, or may be fully or partially integrated into any of the other components in the computer terminal 10 (or mobile device). As in the AR scene-based virtual image generation method involved in the embodiments of the present application, the data processing circuit serves as a processor control (for example, the selection of a variable resistor terminal path connected to an interface).

[0041] The memory 104 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the virtual image generation method based on the AR scene in the embodiment of the present application. The processor 102 executes various functional applications and data processing by running the software programs and modules stored in the memory 104, that is, realizing the above-mentioned virtual image generation method based on the AR scene. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories may be connected to the computer terminal 10 via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0042] The transmission device 106 is used to receive or send data via a network. The specific example of the above network may include a wireless network provided by a communication provider of the computer terminal 10. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0043] The display may be, for example, a touch screen liquid crystal display (LCD) that enables a user to interact with a user interface of the computer terminal 10 (or mobile device).

[0044] Under the above operating environment, this application provides Figure 2 The virtual image generation method based on AR scene is shown. Figure 2FIG. 1 is a flowchart of an optional method for generating a virtual image based on an AR scene according to an embodiment of the present application. Figure 2 As shown, the method comprises the following steps:

[0045] Step S201, obtaining the image style preference information of the target user and the virtual image tag of the target user, wherein the virtual image tag is an image feature of the virtual image corresponding to the target user.

[0046] Optionally, the avatar generation system performs quantitative extraction of the target user's personalized image style information, specifically including: in-depth analysis of the target user's historical interaction records and personal preference data in augmented reality (AR) scenarios. The historical interaction records cover the avatar samples selected by the user in the AR environment, the types of virtual activities participated in, and the content of community interaction, while the personal preference data may come from the self-descriptive text, image materials, or preference settings directly provided by the user.

[0047] Optionally, the avatar tag is a serialized, structured data set specifically used to describe the appearance characteristics, style attributes, and cultural elements that the target user expects his or her avatar to present. The tag system not only covers the description of the overall image style, such as painting style preferences and color choices, but also goes deep into local details, such as clothing style, physical features, and even posture and background environment style.

[0048] Optionally, the target user actively participates in the avatar label setting process, aiming to accurately capture the user's personalized needs. Users can intuitively express their preferences and expectations for avatars through a variety of methods such as text input, voice commands, or image examples. The collection and organization of these labels form a user-defined, multi-dimensional, and flexible avatar description framework, which provides direct guidance for the personalized generation of avatars.

[0049] Step S202: determining N candidate materials from the virtual image material library according to the image style preference information and the virtual image tag, where N is an integer greater than 1.

[0050] Optionally, the avatar generation system performs high-precision intelligent screening of candidate materials from the pre-built avatar material library based on the target user's image style preference information and avatar labels. This process makes full use of machine learning and data mining technology, and through in-depth analysis of user preference information and actively set labels, constructs a multi-dimensional feature vector space, including but not limited to overall image style, local style preferences, painting style attributes and color tendencies.

[0051] Optionally, the virtual image generation system uses advanced matching algorithms to compare and analyze the feature vectors of each material in the material library, thereby determining N highly matching candidate materials, where N is an integer greater than 1, representing a diverse set of materials recommended by the system to the user.

[0052] Optionally, N is set to an integer greater than 1. The selection of N has important scientific basis and practical significance. First, the value of N greater than 1 ensures that the system can recommend multiple alternative materials to users instead of a single choice, thereby providing greater creative freedom and diversity to meet the needs and preferences of different users. Secondly, more than one set of alternative materials provides a rich raw material library for the subsequent generation of virtual image combinations. Through the permutation and combination of algorithms, multiple different virtual image schemes can be generated, and then through user feedback and selection, an image that best meets user expectations is finally determined.

[0053] Step S203: Generate M virtual images according to the N candidate materials, where M is an integer greater than or equal to 1, and the virtual image is composed of at least one candidate material among the N candidate materials.

[0054] Optionally, after determining N candidate materials, the avatar generation system uses an advanced combination generation algorithm to intelligently construct M avatars from these materials. This process includes intelligent analysis of the materials, feature extraction, and multi-dimensional evaluation based on user preferences, ensuring that each generated avatar contains the user's personalized style while meeting the display requirements of the AR scene.

[0055] Optionally, M is set to an integer greater than or equal to 1. The selection of this parameter reflects the application's deep understanding of personalized generation and refined satisfaction of user needs. The setting of the M value not only takes into account the diversity of generated virtual images, ensuring that users can select the image that best meets their expectations from multiple options, but also takes into account the rational use of system resources, avoiding the waste of computing resources and user selection burden caused by generating too many virtual images.

[0056] Optionally, each virtual image generated by the virtual image generation system is composed of at least one of the N candidate materials. Each candidate material, whether it is the skeleton of the overall image, the texture of local features, or accessories of a specific style, is regarded as a module for constructing the virtual image. These modules can flexibly create infinite possible virtual images through the arrangement and combination of intelligent algorithms, which not only ensures the uniqueness and innovation of each image, but also realizes the perfect combination of user personalized needs and AR scene display effects.

[0057] Step S204: adjusting the virtual image according to the scene information of the AR scene to obtain a target virtual image corresponding to the target user.

[0058] Optionally, the virtual image generation system introduces a real-time information feedback mechanism of the AR scene on the basis of generating M virtual images, and dynamically adjusts and optimizes these virtual images in a scientific and precise way. This feedback utilizes the data of environmental perception and user interaction in augmented reality technology, including but not limited to multiple information such as lighting conditions, geographic location, real-time dynamics and user behavior patterns. Through deep learning and computer vision algorithms, the virtual image generation system can intelligently analyze the characteristics of the AR scene, including the background environment, physical conditions and interaction requirements, and integrate this information with the visual presentation and behavior pattern of the virtual image to achieve comprehensive optimization of the virtual image in scene adaptability, visual coordination and interaction fluency. This mechanism not only enhances the natural integration of the virtual image and the AR environment, but also ensures the expressiveness and interactivity of the virtual image in different scenes, providing users with a more immersive and realistic AR experience.

[0059] Optionally, the target virtual image not only fully reflects the user's personalized preferences and style characteristics, but also deeply adapts to the display requirements of specific AR scenes, and has extremely high scene integration and visual appeal.

[0060] Optionally, by combining AR scene information with dynamic adjustment of the virtual image generation process, the virtual image generation system can accurately select a virtual image from M virtual images that best meets the needs of the target user and the current scene display effect, ensuring that the user has a virtual avatar that is both personalized and context-adaptive in the AR environment.

[0061] It can be seen from the contents of steps S201 to S204 that in an embodiment of the present application, the image style preference information of the target user and the virtual image label of the target user are first obtained, wherein the virtual image label is the image feature of the virtual image corresponding to the target user's description, and then according to the image style preference information and the virtual image label, N alternative materials are determined from the virtual image material library, wherein N is an integer greater than 1, and then M virtual images are generated according to the N alternative materials, wherein M is an integer greater than or equal to 1, and the virtual image is composed of at least one alternative material among the N alternative materials, and finally the virtual image is adjusted according to the scene information of the AR scene to obtain the target virtual image corresponding to the target user.

[0062] From the above content, it can be seen that this application comprehensively quantifies the user's style preferences by analyzing the user's historical behavior data in the AR scene, such as the virtual image selected by the user, the activities participated in, and the interactive information in the community. In addition, the user can actively describe the characteristics of the virtual image he or she expects and set labels for the virtual image.

[0063] Secondly, based on the style preference information and virtual image labels obtained in the previous step, N alternative materials are intelligently selected from the pre-classified virtual image material library, where N is an integer greater than 1. These materials cover multiple dimensions of user preferences, such as overall image style, local style, painting style and color information.

[0064] In addition, M different virtual images are generated using the selected N candidate materials, where M is an integer greater than or equal to 1. Each generated virtual image is composed of at least one candidate material, ensuring that the generated virtual image can fully reflect the personalized needs of the user.

[0065] Finally, the generated virtual image is adjusted to adapt to the scene based on the real-time information of the target user's AR scene, such as the real environment, lighting conditions, etc. This includes not only adjusting the scale and position of the virtual image to match the user's real environment, but also adjusting the material properties and lighting effects to enhance the visual integration of the virtual image and the AR scene.

[0066] It can be seen that through the technical solution of the present application, the purpose of achieving a deep understanding of user style preferences is achieved, thereby achieving the technical effect of ensuring that the generated virtual image meets personalized needs and enhancing the user's immersive experience and interactive fun in the AR environment, thereby solving the technical problem that the existing technology cannot generate personalized virtual images in AR scenes.

[0067] In an optional embodiment, the virtual image generation system first obtains the historical behavior information of the target user in the AR scene, wherein the historical behavior information at least includes the virtual images selected by the target user in the AR scene, the activity information participated in by the target user in the AR scene, and the social interaction information about the virtual image published by the target user in the AR scene, and then determines the image style preference information of the target user from the historical behavior information.

[0068] Optionally, the in-depth mining and analysis of the target user's historical behavior data by the avatar generation system mainly includes in-depth mining of the user's multi-dimensional behavior data in the AR environment, including but not limited to the avatars selected by the user, the information of the activities participated in, and the interactive communication records around the avatar in the community. By building a comprehensive historical behavior information database, this application provides a rich data foundation for the subsequent determination of image style preference information.

[0069] Optionally, the historical behavior information includes a series of key interactive behaviors of the target user in the AR scene, among which the selected virtual images reflect the user's preferences for different styles and features; the information on the activities participated in reveals the user's behavior patterns and interests in different situations; and the community interaction information provides the user's social preferences and emotional tendencies in the virtual community, including discussions, evaluations, and recommendations for specific virtual images. These diverse historical behavior data together constitute the user's personalized behavior portrait in the AR scene, providing in-depth insights and basis for the customization of virtual images.

[0070] Optionally, the image style preference information is determined mainly by scientifically determining the target user's specific image style preference information from the above historical behavior information through advanced data mining and machine learning algorithms. The process of determining the image style preference information by the virtual image generation system involves cleaning, preprocessing, feature extraction and pattern recognition of historical data. By building a user preference model, it can accurately identify the user's personalized characteristics such as style tendencies, emotional attributes, cultural elements, etc. for the virtual image. Based on the preference information obtained from these analyses, the virtual image generation system can more accurately screen alternative materials from the virtual image material library and customize images that highly meet the user's personalized needs, thereby greatly enhancing the user's immersion and interactive experience in the AR scene.

[0071] From the above content, we can see that the virtual image generation system deeply mines historical behavior information and uses scientific algorithms to determine the user's specific image style preferences. It not only provides accurate data support for the generation of personalized virtual images, but also reflects a deep understanding and respect for users' personalized needs.

[0072] In an optional embodiment, the virtual image generation system first obtains a target text and a target image submitted by a target user, wherein the target text records the target user's descriptive information about the desired virtual image, and the target image includes the virtual image that the target user expects to use for reference, and then extracts the target user's virtual image label from the target text and the target image.

[0073] Optionally, the virtual image generation system first scientifically integrates the target text and target image data submitted by the target user as an important data source that directly reflects the user's personalized needs, emphasizes the accurate acquisition and systematic processing of user data, and ensures the reliability and effectiveness of subsequent analysis.

[0074] Optionally, in the target text, the user describes his or her expectations for the characteristics of the virtual image in natural language, and these descriptions include preferences for colors, styles, elements, and expected performance of the virtual image in the AR scene.

[0075] Optionally, the target image serves as an intuitive display of the user's ideal virtual image and contains rich visual information.

[0076] Optionally, the semantic features and visual information extracted by the avatar generation system from the target text and target image are then converted into a series of structured, semantically clear and emotionally rich avatar labels through an algorithm. These labels not only cover the user's direct description of the avatar's appearance, such as "secondary style" and "future warrior equipment", but also include the user's inherent attributes and emotional characteristics that the avatar should show in the AR scene, such as "bravery" and "wisdom". Through the label generation process, the avatar generation system can build a description framework that fully reflects the user's personalized needs and emotional preferences, providing a multi-dimensional guiding basis for the customized generation of avatars, ensuring that the generated avatars meet user expectations in appearance.

[0077] From the above content, we can see that the virtual image generation system extracts personalized tags of the virtual image by scientifically analyzing the target text and target image submitted by the user. This not only deepens the system's understanding of user needs, but also provides precise guidance for the generation of virtual images, promoting the innovative development of personalized virtual images in AR technology.

[0078] In an optional embodiment, the virtual image generation system first generates a preference feature label of the target user based on the image style preference information, and then merges the preference feature label and the virtual image label into an overall image style label of the target user, and determines N alternative materials from the virtual image material library based on the overall image style label, wherein the overall image style label includes at least one of the following dimensions: overall image style information, local style information, painting style information, and color information.

[0079] Optionally, the avatar generation system uses scientific algorithms to convert the image style preference information parsed from the user's historical behavior information in the AR scene into a series of structured and semantically clear preference feature labels. These preference feature labels contain the user's intuitive description and implicit needs for the avatar, such as "cute" and "sense of technology", which are a direct reflection of the user's personalized preferences at the data and feature level. Through deep learning and data mining technology, the avatar generation system can accurately capture the user's style tendencies and provide a key basis for subsequent label fusion and selection of avatar materials.

[0080] Optionally, after obtaining the preferred feature labels and the avatar labels directly extracted from the user's submitted information, the avatar generation system uses advanced label fusion technology to deeply integrate the two in multiple dimensions to construct the overall image style label of the target user. The avatar generation system comprehensively considers the user's historical behavior patterns and immediate description preferences, ensuring that the final generated overall image style label can fully reflect the user's avatar needs in the AR scene, not only visually, but also highly matched at the emotional and cultural levels.

[0081] Optionally, based on the constructed user's overall image style label, the avatar generation system intelligently selects N candidate materials from the avatar material library. Based on the feature matching and scoring of the label, through multi-dimensional comparison and comprehensive evaluation, the avatar generation system can accurately identify which materials best match the user's overall style preference. The selection of N candidate materials ensures the richness of materials and style diversity of subsequent avatar generation, providing a solid foundation for the customization of personalized avatars.

[0082] Optionally, the overall image style label covers at least one of the following dimensions: overall image style information (such as "ancient style", "modern city"), local style information (such as "Gothic clothing", "natural style hairstyle"), painting style information (such as "cartoon style", "realistic style"), and color information (such as "warm tones", "cold tones"). By constructing labels in multiple dimensions, the virtual image generation system can fully capture the user's style preferences, ensuring that the generated virtual image not only meets the user's needs in terms of overall style, but also achieves personalization in local details, visual presentation and color matching, greatly improving the visual appeal and emotional communication ability of the virtual image in the AR scene.

[0083] From the above content, it can be seen that the virtual image generation system realizes the intelligent screening of virtual image materials based on the overall image style label through the feature conversion, label fusion and multi-dimensional construction of user style preference information. It not only accurately reflects the user's personalized needs for virtual images in AR scenes, but also provides scientific guidance for the customized generation of virtual images.

[0084] In an optional embodiment, the virtual image generation system first obtains a matching score of each virtual image material in a virtual image material library on each dimension of the overall image style label, wherein the virtual image material library includes pre-classified virtual image materials of multiple styles, and then performs feature matching with the overall image style label on each dimension for the same type of material according to the matching score, and screens out X groups of virtual image materials in each dimension that meet the requirements of the overall image style label, wherein X is an integer greater than 1, and then takes an intersection of the screened X groups of virtual image materials to obtain a candidate material set consisting of N candidate materials.

[0085] Optionally, during the process of generating personalized virtual images, the virtual image generation system scientifically calculates the matching score of each material in the virtual image material library with the overall image style label of the target user in each dimension. This step uses the deep similarity matching algorithm and feature vector comparison technology to quantitatively analyze the multi-dimensional features of each material, such as style attributes, local elements, painting style characteristics and color tendencies, and accurately compares them with the user's overall style label, thereby obtaining the matching score of each material in each dimension, providing a quantitative basis for subsequent material screening.

[0086] Optionally, the avatar material library contains pre-classified avatar materials of various styles, which cover a wide range of styles from traditional to modern, from abstract to concrete, from cartoon to realistic, etc., ensuring the diversity and richness of the materials. The pre-classification of the material library is based on professional design principles and style classification standards, which facilitates rapid retrieval and accurate matching based on the user's overall image style label in the subsequent feature matching stage.

[0087] Optionally, the avatar generation system systematically matches the features of the same type of material in different dimensions based on the calculated matching score, and selects X groups (X is an integer greater than 1) of avatar materials that highly match the overall style label of the target user. The screening process fully considers the complexity and compatibility of user preferences in different dimensions, ensuring that each group of materials is highly matched with user preferences in multiple dimensions such as overall image style, local style, painting style and color. The selection of X groups of materials not only enriches the search results of the material library, but also provides a variety of material combination possibilities, providing flexibility and innovation space for the subsequent generation of personalized avatars.

[0088] Optionally, the virtual image generation system performs a scientific intersection operation on the selected X groups of virtual image materials to generate a candidate material set consisting of N candidate materials. This process involves identifying and integrating the common features of the X groups of materials in different dimensions, ensuring that each material in the set meets the overall style preferences of the target user in all dimensions, thereby achieving the optimization of the material set. The determination of N candidate materials provides an accurate and rich material library for the subsequent generation of virtual images, ensuring the personalization and accuracy of the generated results in visual presentation and emotional expression.

[0089] From the above content, it can be seen that the virtual image generation system not only accurately reflects the target users' personalized needs for virtual images in AR scenarios through multi-dimensional matching evaluation of virtual image materials, screening and combination of multiple groups of materials based on matching scores, and intelligent generation and optimization of alternative material sets, but also greatly enriches the flexibility and innovation space of material selection, providing scientific guidance for the customized generation of personalized virtual images.

[0090] In an optional embodiment, the virtual image generation system first obtains the real environment information of the target user through the camera and sensor of the AR device, and adjusts the scale and position of the virtual image according to the real environment information of the target user, and then and / or obtains the lighting conditions in the AR scene, and adjusts the material properties and lighting position of the virtual image according to the lighting conditions.

[0091] Optionally, the virtual image generation system uses scientific methods to capture the real environment information of the target user in real time through the high-precision camera and sensors of the AR device. This environmental perception process involves the collection and analysis of multi-dimensional data such as spatial positioning, ambient lighting, and scene characteristics, providing real-world data support for the subsequent proportion adjustment and position layout of the virtual image. Through deep learning and environmental recognition technology, the virtual image generation system can accurately understand the complexity of the user's surrounding environment and provide accurate background reference for the presentation of the virtual image.

[0092] Optionally, based on the acquired real environment information of the target user, the virtual image generation system uses scientific algorithms to dynamically adjust the proportion and position of the virtual image to ensure seamless integration of the virtual image with the real environment. The virtual image generation system fully considers the user's perspective, spatial scale, and scene layout. By intelligently calculating the relative position and proportion relationship between the virtual image and the surrounding objects, it can achieve the natural embedding of the virtual image in the real space, thereby enhancing the realism and immersion of the AR experience.

[0093] Optionally, the virtual image generation system also scientifically utilizes the lighting conditions in the AR scene, collects ambient lighting data through high-precision sensors, including light intensity, direction, and color temperature, to provide accurate adjustment basis for the material properties and lighting effects of the virtual image. Based on the acquired lighting conditions, the virtual image generation system uses physical rendering and lighting simulation algorithms to adjust the material properties of the virtual image, such as color saturation, surface reflection, and shadow distribution, as well as the position and intensity of the light source, to ensure that the virtual image can show the best visual effect under different lighting conditions, in harmony with the lighting conditions of the real environment, and enhance the visual authenticity and interactive experience of the AR scene.

[0094] From the above content, we can see that the virtual image generation system uses the camera and sensors of the AR device to scientifically collect and analyze the real environment information of the target user, and uses intelligent algorithms to dynamically adjust the proportion, position, material and lighting of the virtual image, which not only ensures the harmonious integration of the virtual image and the real environment, but also greatly enhances the realism and immersion of the AR experience.

[0095] In an optional embodiment, the image style preference information includes at least one of the following style preference information: clothing style preference information of the virtual image; appearance style preference information of the virtual image; background environment style preference information of the virtual image; and posture style preference information of the virtual image.

[0096] Optionally, the avatar generation system first scientifically analyzes the target user's preference information on clothing style in the customization of personalized avatars. The avatar generation system extracts multi-dimensional preference features such as clothing style attributes, color preferences, material selection, and pattern design from the user's input data through image recognition technology, semantic analysis, and data mining methods. These preference information is quantified into structured data labels, such as "retro", "technological future", "natural pastoral", etc., which provides clear guidance for the subsequent design and selection of avatar clothing.

[0097] Optionally, physical features are an important component of the personalized presentation of the avatar. The avatar generation system uses computer vision technology, especially facial recognition and posture analysis algorithms, to accurately identify the target user's preference information on physical style, including facial features, body proportions, posture movements, etc. Through in-depth analysis of user description information and reference images, the avatar generation system can build a preference model of the target user's physical features, providing a scientific basis for the design of the avatar's physical style, ensuring that the generated avatar is highly consistent with the user's preferences in terms of physical appearance.

[0098] Optionally, the background environment plays a vital role in the presentation effect and emotional communication of the virtual image. The virtual image generation system uses the environmental perception technology of the AR device, combined with the environmental style preferences input by the user, to scientifically analyze and identify the user's style requirements for the background environment in the AR scene, including but not limited to natural scenery, urban landscape, future world, etc. Through a deep understanding of the background environment style preference information, the virtual image generation system can seamlessly integrate the user's preferred environmental style with the design of the virtual image, creating a virtual image that not only meets the user's personalized needs but also coordinates with the AR scene background, thereby enhancing the user's immersion and participation in the AR experience.

[0099] Optionally, the dynamic posture of the avatar is a key factor in enhancing its realism and vividness in the AR scene. The avatar generation system combines motion capture technology with machine learning algorithms to accurately identify and quantify the target user's preference information for the avatar's posture style, including standing posture, walking mode, expression changes, etc. Through the scientific capture and analysis of user posture preferences, the avatar generation system can simulate the dynamic performance of the avatar that meets the user's expectations, ensuring that the avatar's movements in the AR scene are natural and smooth, and showing rich expressions and posture changes when interacting with users, greatly enhancing the vividness and interactivity of the AR experience.

[0100] From the above content, it can be seen that the virtual image generation system not only ensures the personalization and accuracy of the virtual image in visual presentation and emotional communication through scientific analysis and quantification of clothing style, body style, background environment style and posture style preference information, but also provides scientific guidance for the dynamic performance and environmental integration of the virtual image in AR scenes.

[0101] According to another aspect of the embodiment of the present application, a virtual image generation device based on an AR scene is also provided, wherein: Figure 3 is a schematic diagram of an optional AR scene-based virtual image generation device according to an embodiment of the present application, such as Figure 3 As shown, the virtual image generation device based on the AR scene includes: an acquisition unit 301, a determination unit 302, a generation unit 303, and an adjustment unit 304.

[0102] An acquisition unit 301 acquires the image style preference information of a target user and a virtual image tag of the target user, wherein the virtual image tag is an image feature of a virtual image corresponding to the target user. A determination unit 302 determines N candidate materials from a virtual image material library according to the image style preference information and the virtual image tag, wherein N is an integer greater than 1. A generation unit 303 generates M virtual images according to the N candidate materials, wherein M is an integer greater than or equal to 1, and the virtual image is composed of at least one of the N candidate materials. An adjustment unit 304 adjusts the virtual image according to the scene information of the AR scene to obtain a target virtual image corresponding to the target user.

[0103] Optionally, the acquisition unit 301 includes: a first acquisition subunit and a first determination subunit. The first acquisition subunit is used to acquire the historical behavior information of the target user in the AR scene, wherein the historical behavior information at least includes the virtual image selected by the target user in the AR scene, the activity information participated by the target user in the AR scene, and the social interaction information about the virtual image published by the target user in the AR scene; the first determination subunit is used to determine the image style preference information of the target user from the historical behavior information.

[0104] Optionally, the acquisition unit 301 includes: a second acquisition subunit and a first extraction subunit. The second acquisition subunit is used to acquire a target text and a target image submitted by a target user, wherein the target text records the description information of the target user on the desired virtual image, and the target image includes the virtual image that the target user expects to use for reference; and the first extraction subunit is used to extract the virtual image label of the target user from the target text and the target image.

[0105] Optionally, the determination unit 302 includes: a first generation subunit and a second determination subunit. The first generation subunit is used to generate a preference feature label of the target user according to the image style preference information; the second determination subunit is used to merge the preference feature label and the virtual image label into an overall image style label of the target user, and determine N candidate materials from the virtual image material library according to the overall image style label, wherein the overall image style label includes at least one of the following dimensions: overall image style information, local style information, painting style information, and color information.

[0106] Optionally, the second determination subunit includes: a first acquisition module, a first processing module, and a second processing module. The first acquisition module is used to obtain the matching score of each virtual image material in the virtual image material library in each dimension of the overall image style label, wherein the virtual image material library includes virtual image materials of multiple styles pre-classified; the first processing module is used to perform feature matching with the overall image style label in each dimension for the same type of material according to the matching score, and screen out X groups of virtual image materials in each dimension that meet the overall image style label, wherein X is an integer greater than 1; the second processing module is used to take the intersection of the screened X groups of virtual image materials to obtain a candidate material set consisting of N candidate materials.

[0107] Optionally, the adjustment unit 304 includes: a third acquisition subunit and a fourth acquisition subunit. The third acquisition subunit is used to acquire the real environment information of the target user through the camera and sensor of the AR device, and adjust the proportion and position of the virtual image according to the real environment information of the target user; the fourth acquisition subunit is used to acquire the lighting conditions in the AR scene, and adjust the material properties and lighting position of the virtual image according to the lighting conditions.

[0108] Optionally, the image style preference information includes at least one of the following style preference information: clothing style preference information of the virtual image; physical appearance style preference information of the virtual image; background environment style preference information of the virtual image; and posture style preference information of the virtual image.

[0109] An embodiment of the present application may provide an electronic device, Figure 4is a structural block diagram of an electronic device according to an embodiment of the present application. Figure 4 As shown, the electronic device may include: one or more ( Figure 4 Only one is shown) a processor, a memory, a storage controller, and a peripheral interface, wherein the peripheral interface is connected to a radio frequency module, an audio module, and a display.

[0110] Among them, the memory can be used to store software programs and modules, such as program instructions / modules corresponding to the methods and devices in the embodiments of the present application, and the processor executes various functional applications and data processing by running the software programs and modules stored in the memory, that is, realizing the above-mentioned method. The memory may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory may further include a memory remotely arranged relative to the processor, and these remote memories may be connected to the terminal via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0111] The processor can call the information and application stored in the memory through the transmission device to perform the following steps: obtain the image style preference information of the target user and the virtual image label of the target user, wherein the virtual image label is the image characteristics of the corresponding virtual image described by the target user; determine N alternative materials from the virtual image material library according to the image style preference information and the virtual image label, wherein N is an integer greater than 1; generate M virtual images according to the N alternative materials, wherein M is an integer greater than or equal to 1, and the virtual image is composed of at least one alternative material among the N alternative materials; adjust the virtual image according to the scene information of the AR scene to obtain the target virtual image corresponding to the target user.

[0112] The processor can also call the information and application programs stored in the memory through the transmission device to perform the following steps: obtain the historical behavior information of the target user in the AR scene, wherein the historical behavior information at least includes the virtual image selected by the target user in the AR scene, the activity information participated in by the target user in the AR scene, and the social interaction information about the virtual image published by the target user in the AR scene; determine the image style preference information of the target user from the historical behavior information.

[0113] The processor can also call the information and application stored in the memory through the transmission device to perform the following steps: obtain the target text and target image submitted by the target user, wherein the target text records the target user's description information about the desired virtual image, and the target image includes the virtual image that the target user expects to use for reference; extract the target user's virtual image label from the target text and target image.

[0114] The processor can also call the information and application programs stored in the memory through the transmission device to perform the following steps: generate a preference feature label of the target user based on the image style preference information; merge the preference feature label and the virtual image label into an overall image style label of the target user, and determine N alternative materials from the virtual image material library based on the overall image style label, wherein the overall image style label includes at least one of the following dimensions: overall image style information, local style information, painting style information, and color information.

[0115] The processor may also call the information and application program stored in the memory through the transmission device to perform the following steps: obtaining a matching score of each virtual image material in the virtual image material library in each dimension of the overall image style label, wherein the virtual image material library includes virtual image materials of multiple styles that are pre-classified; performing feature matching with the overall image style label in each dimension for the same type of material according to the matching score, and screening out X groups of virtual image materials in each dimension that meet the overall image style label, wherein X is an integer greater than 1; and taking an intersection of the screened X groups of virtual image materials to obtain a candidate material set consisting of N candidate materials.

[0116] The processor can also call the information and application programs stored in the memory through the transmission device to perform the following steps: obtain the real environment information of the target user through the camera and sensor of the AR device, and adjust the scale and position of the virtual image according to the real environment information of the target user; and / or obtain the lighting conditions in the AR scene, and adjust the material properties and lighting position of the virtual image according to the lighting conditions.

[0117] The processor can also call the information and application programs stored in the memory through the transmission device to perform the following steps: the image style preference information includes at least one of the following style preference information: the clothing style preference information of the virtual image; the appearance style preference information of the virtual image; the background environment style preference information of the virtual image; the posture style preference information of the virtual image.

[0118] In an embodiment of the present application, firstly, the image style preference information of the target user and the virtual image label of the target user are obtained, wherein the virtual image label is the image feature of the virtual image corresponding to the target user's description, and then, based on the image style preference information and the virtual image label, N candidate materials are determined from the virtual image material library, wherein N is an integer greater than 1, and then, based on the N candidate materials, M virtual images are generated, wherein M is an integer greater than or equal to 1, and the virtual image is composed of at least one candidate material among the N candidate materials, and finally, the virtual image is adjusted according to the scene information of the AR scene to obtain the target virtual image corresponding to the target user.

[0119] From the above content, it can be seen that this application comprehensively quantifies the user's style preferences by analyzing the user's historical behavior data in the AR scene, such as the virtual image selected by the user, the activities participated in, and the interactive information in the community. In addition, the user can actively describe the characteristics of the virtual image he or she expects and set labels for the virtual image.

[0120] Secondly, based on the style preference information and virtual image labels obtained in the previous step, N alternative materials are intelligently selected from the pre-classified virtual image material library, where N is an integer greater than 1. These materials cover multiple dimensions of user preferences, such as overall image style, local style, painting style and color information.

[0121] In addition, M different virtual images are generated using the selected N candidate materials, where M is an integer greater than or equal to 1. Each generated virtual image is composed of at least one candidate material, ensuring that the generated virtual image can fully reflect the personalized needs of the user.

[0122] Finally, the generated virtual image is adjusted to adapt to the scene based on the real-time information of the target user's AR scene, such as the real environment, lighting conditions, etc. This includes not only adjusting the scale and position of the virtual image to match the user's real environment, but also adjusting the material properties and lighting effects to enhance the visual integration of the virtual image and the AR scene.

[0123] It can be seen that through the technical solution of the present application, the purpose of achieving a deep understanding of user style preferences is achieved, thereby achieving the technical effect of ensuring that the generated virtual image meets personalized needs and enhancing the user's immersive experience and interactive fun in the AR environment, thereby solving the technical problem that the existing technology cannot generate personalized virtual images in AR scenes.

[0124] It can be understood by those skilled in the art that Figure 4 The structure shown is for illustration only, and the electronic device may also be a smart phone (such as an Android phone, an iOS phone, etc.), a tablet computer, a PDA, a mobile Internet device (Mobile Internet Devices, MID), a PAD, or other terminal devices. Figure 4 The structure of the electronic device is not limited. Figure 4 More or fewer components (such as network interfaces, display devices, etc.) shown in, or having Figure 4 Different configurations shown.

[0125] A person of ordinary skill in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing the hardware related to the terminal device through a program, and the program can be stored in a computer-readable storage medium, and the storage medium may include: a flash drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0126] According to another aspect of the present application, a computer-readable storage medium is also provided, wherein a computer program is stored in the computer-readable storage medium, wherein when the computer program is executed, the device where the computer-readable storage medium is located executes the above-mentioned AR scene-based virtual image generation method.

[0127] According to another aspect of the present application, a computer program product is also provided, wherein the computer program product includes computer instructions, wherein when the computer instructions are executed, the device where the computer program product is located executes the above-mentioned virtual image generation method based on AR scene.

[0128] The above-mentioned embodiments or examples disclosed in the present application are not exhaustive, but are only illustrative of some embodiments or examples, and are not intended to be specific limitations on the scope of protection disclosed in the present application. In the absence of contradiction, each step in a certain embodiment or example in the present application can be implemented as an independent example, and the steps can be combined arbitrarily. For example, the scheme after removing some steps in a certain embodiment or example can also be implemented as an independent example, and the order of the steps in a certain embodiment or example can be arbitrarily exchanged. In addition, the optional methods or optional examples in a certain embodiment or example can be combined arbitrarily; in addition, the various embodiments or examples can be combined arbitrarily, for example, some or all steps of different embodiments or examples can be combined arbitrarily, and a certain embodiment or example can be combined arbitrarily with the optional methods or optional examples of other embodiments or examples.

[0129] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0130] In the above embodiments of the present application, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0131] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units can be a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0132] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0133] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0134] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, disk or optical disk and other media that can store program codes.

[0135] The above is only a preferred implementation of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A method for generating a virtual image based on an AR scene, characterized in that: include: Acquire the image style preference information of the target user and the virtual image label of the target user, wherein the virtual image label is the image feature of the virtual image corresponding to the target user; Determining N candidate materials from a virtual image material library according to the image style preference information and the virtual image tag, where N is an integer greater than 1; Generate M virtual images according to the N candidate materials, where M is an integer greater than or equal to 1, and the virtual image is composed of at least one candidate material among the N candidate materials; The virtual image is adjusted according to the scene information of the AR scene to obtain a target virtual image corresponding to the target user.

2. The method for generating a virtual image based on an AR scene according to claim 1, characterized in that: Obtain the target user's image style preference information, including: Acquire historical behavior information of the target user in the AR scene, wherein the historical behavior information at least includes virtual images selected by the target user in the AR scene, information about activities participated by the target user in the AR scene, and social interaction information about the virtual images published by the target user in the AR scene; The image style preference information of the target user is determined from the historical behavior information.

3. The method for generating a virtual image based on an AR scene according to claim 1, characterized in that: Get the target user's avatar tags, including: Acquire a target text and a target image submitted by the target user, wherein the target text records description information of the target user about the desired virtual image, and the target image includes the virtual image that the target user desires to use for reference; A virtual image label of the target user is extracted from the target text and the target image.

4. The method for generating a virtual image based on an AR scene according to claim 1, characterized in that: According to the image style preference information and the virtual image tag, N candidate materials are determined from the virtual image material library, including: Generating a preference feature label of the target user according to the image style preference information; The preference feature label and the virtual image label are merged into an overall image style label of the target user, and the N candidate materials are determined from the virtual image material library according to the overall image style label, wherein the overall image style label includes at least one of the following dimensions: Overall image style information, local style information, painting style information, and color information.

5. The method for generating a virtual image based on an AR scene according to claim 4, characterized in that: Determining the N candidate materials from the virtual image material library according to the overall image style label includes: Obtaining a matching score of each virtual image material in the virtual image material library on each dimension of the overall image style tag, wherein the virtual image material library includes virtual image materials of multiple styles that are pre-classified; According to the matching score, for the same type of material, feature matching is performed with the overall image style label in each dimension, and X groups of virtual image materials that meet the dimensions of the overall image style label are screened out, where X is an integer greater than 1; The selected X groups of virtual image materials are intersected to obtain a candidate material set consisting of N candidate materials.

6. The method for generating a virtual image based on an AR scene according to claim 1, characterized in that: The virtual image is adjusted according to the scene information of the AR scene, including: Acquire the real environment information of the target user through the camera and sensor of the AR device, and adjust the scale and position of the virtual image according to the real environment information of the target user; and / or, Acquire lighting conditions in the AR scene, and adjust material properties and lighting position of the virtual image according to the lighting conditions.

7. The method for generating a virtual image based on an AR scene according to claim 1, characterized in that: The image style preference information includes at least one of the following style preference information: Clothing style preference information of the virtual image; Physical appearance style preference information of the virtual image; Background environment style preference information of the virtual image; The posture style preference information of the virtual image.

8. A virtual image generation device based on AR scene, characterized in that: include: An acquisition unit, which acquires the image style preference information of the target user and the virtual image tag of the target user, wherein the virtual image tag is an image feature of the virtual image corresponding to the target user; a determining unit, which determines N candidate materials from a virtual image material library according to the image style preference information and the virtual image tag, wherein N is an integer greater than 1; a generating unit, generating M virtual images according to the N candidate materials, wherein M is an integer greater than or equal to 1, and the virtual image is composed of at least one candidate material among the N candidate materials; The adjustment unit adjusts the virtual image according to the scene information of the AR scene to obtain a target virtual image corresponding to the target user.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored executable program, wherein when the executable program is running, the device where the computer-readable storage medium is located is controlled to execute the AR scene-based virtual image generation method according to any one of claims 1 to 7.

10. An electronic device, characterized in that: include: A memory storing an executable program; A processor is used to run the program, wherein when the program is run, the method for generating a virtual image based on an AR scene according to any one of claims 1 to 7 is executed.

11. A computer program product comprising computer instructions, characterized in that: When the computer instructions are executed by the processor, the steps of the method for generating a virtual image based on an AR scene according to any one of claims 1 to 7 are implemented.

Images