Group photo method, device and equipment of virtual images, medium and product

Abstract

The invention provides a group photo method and device for virtual images, equipment, a medium and a product, and relates to the technical fields of large models, generative forms, generative models, Transformers, computer graphics, artificial intelligence, virtual images, image synthesis, cloud computing and the like. According to the specific implementation scheme, the method comprises the steps of determining a current image feature corresponding to a current object and other image features corresponding to other objects except the current object; generating a current virtual image prototype corresponding to the current object according to the current image feature, and generating other virtual image prototypes corresponding to other objects according to other image features; and generating a virtual image group photo between the current object and other objects according to the current virtual image prototype and other virtual image prototypes. According to the method and the device, the anthropomorphic effect of the virtual image group photo is improved, so that the virtual image group photo is closer to the real image of each object participating in the group photo, and the use experience of a user is improved.

Description

Technical Field

[0001] The present disclosure relates to the field of computer technology, specifically to technical fields such as large models, generative formulas, generative models, Transformers, computer graphics, artificial intelligence, virtual images, image synthesis and cloud computing, and more particularly to a method, device, equipment, medium and product for taking group photos with a virtual image. Background Art

[0002] With the continuous development of Internet technology, avatar technology has also developed rapidly. Avatar is a virtual character created and designed by digital means based on computer graphics and artificial intelligence technology. It is precisely because of the rapid development of avatar technology that users are not only satisfied with creating their own avatars, but also hope to take photos of their own avatars with other users' avatars.

[0003] The current virtual image photo-taking technology usually replaces the face based on the real user’s photo according to the facial features of different users to achieve the virtual image photo-taking effect, commonly known as "AI face-changing". Summary of the invention

[0004] The present disclosure provides a virtual image group photo taking method, device, equipment, medium and product for improving the anthropomorphism of the virtual image group photo.

[0005] According to one aspect of the present disclosure, a method for taking a group photo of a virtual image is provided, comprising:

[0006] Determine a current image feature corresponding to a current object, and other image features corresponding to other objects except the current object;

[0007] Generating a current virtual image prototype corresponding to the current object according to the current image feature, and generating other virtual image prototypes corresponding to the other objects according to the other image features;

[0008] A virtual image photo of the current object and the other objects is generated according to the current virtual image prototype and the other virtual image prototypes.

[0009] According to another aspect of the present disclosure, a device for taking a group photo of a virtual image is provided, comprising:

[0010] An image feature determination module, used to determine a current image feature corresponding to a current object, and other image features corresponding to other objects except the current object;

[0011] A virtual image prototype generation module, used to generate a current virtual image prototype corresponding to the current object according to the current image feature, and to generate other virtual image prototypes corresponding to the other objects according to the other image features;

[0012] The virtual image group photo generation module is used to generate a virtual image group photo between the current object and the other objects according to the current virtual image prototype and the other virtual image prototypes.

[0013] According to another aspect of the present disclosure, there is provided an electronic device, comprising:

[0014] at least one processor; and

[0015] a memory communicatively connected to at least one processor; wherein,

[0016] The memory stores instructions that can be executed by at least one processor. The instructions are executed by at least one processor to enable the at least one processor to perform any method in the present disclosure.

[0017] According to another aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to cause a computer to execute any one of the methods in the present disclosure.

[0018] According to another aspect of the present disclosure, a computer program product is provided, including a computer program, and the computer program executes any one of the methods of the present disclosure when a processor is used.

[0019] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The accompanying drawings are used to better understand the present solution and do not constitute a limitation of the present disclosure.

[0021] Figure 1 is a flow chart of some virtual image group photo taking methods disclosed in embodiments of the present disclosure;

[0022] Figure 2A is a flow chart of another method for taking a group photo of virtual images disclosed in an embodiment of the present disclosure;

[0023] Figure 2B is a schematic diagram of some current image feature configuration pages disclosed according to an embodiment of the present disclosure;

[0024] Figure 3 is a flow chart of another method for taking a group photo of virtual images disclosed in an embodiment of the present disclosure;

[0025] Figure 4 It is a schematic diagram of the structure of some virtual image group photo taking devices disclosed in the embodiments of the present disclosure;

[0026] Figure 5 It is a block diagram of an electronic device used to implement the method for taking a group photo with a virtual image disclosed in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0027] The following is a description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0028] The current virtual image photo-taking technology usually replaces the face based on the real user’s photo according to the facial features of different users to achieve the virtual image photo-taking effect, commonly known as "AI face-changing".

[0029] However, this method can only ensure the restoration of facial features in virtual image photos, while other image features cannot be close to the real images of the subjects participating in the photo, resulting in poor anthropomorphism of virtual image photos and poor user experience.

[0030] Figure 1 The present invention is a flowchart of some virtual image group photo methods disclosed in the embodiments of the present invention. The present embodiment can be applied to the case of generating a virtual image group photo between the current object and other objects. The method of the present embodiment can be executed by the virtual image group photo device disclosed in the embodiments of the present invention. The device can be implemented by software and / or hardware and can be integrated in any electronic device with computing capabilities, such as a computer.

[0031] like Figure 1 As shown, the method for taking a group photo of a virtual image disclosed in this embodiment may include:

[0032] S101, determining a current image feature corresponding to a current object, and other image features corresponding to other objects except the current object.

[0033] Among them, the current object can be either a "person" or a "non-person entity", and correspondingly, other objects can be either a "person" or a "non-person entity".

[0034] Optionally, the current object may be the main user who currently controls the virtual image photo taking software, that is, the user who has the need for virtual image photo taking and has been authorized to control the virtual image photo taking software; the current object may also be a non-human entity that the main user expects to take a virtual image photo with, such as an animal or a toy model, etc. Correspondingly, other objects may be guest users invited by the main user to take a virtual image photo, that is, the user who has the need for virtual image photo taking and is invited by the main user to take a virtual image photo; other objects may also be non-human entities that the guest user expects to take a virtual image photo with, such as an animal or a toy model, etc. Among them, the type of the virtual image photo taking software may be an APP (Application) installed in a smart terminal, or a mini-program in an APP, etc.

[0035] Image features refer to the characteristics of an object's external appearance, such as appearance and demeanor, and are the overall impression formed in the brain through sensory organs such as vision, hearing, touch and taste. Taking the object as a "person" as an example, the image features include but are not limited to age features, gender features, appearance features, style features, etc. It can be understood that the current image features refer to the image features of the current object; other image features refer to the image features of other objects.

[0036] Determining the current image feature corresponding to the current object includes but is not limited to the following implementations:

[0037] In one implementation, the current subject logs in to a virtual image photo taking software (hereinafter referred to as the target software), and implements an image feature configuration request on the target software. The target software generates a current image feature configuration page corresponding to the current subject according to the image feature configuration request. The current image feature configuration page displays at least one type of image feature to be configured, and the current subject implements configuration operations on each type of image feature to be configured according to its own real image feature. The target software determines the current image feature corresponding to the current subject according to the configuration operations implemented by the current subject on each type of image feature to be configured.

[0038] In another embodiment, the current image feature configuration page displays at least one type of image feature to be configured, and candidate image features corresponding to each type of image feature to be configured. The current object selects features from the candidate image features corresponding to each type of image feature to be configured based on its own real image features. The target software determines the current image feature corresponding to the current object based on the selection operation of the current object on each candidate image feature. For example, the current object selects "youth" in "age feature", selects "black long straight hair" in "appearance feature", and selects "simple style" in "style feature", and the target software correspondingly determines that the current image features corresponding to the current object are "youth", "black long straight hair" and "simple style".

[0039] In another embodiment, the current image feature configuration page includes an image upload button. If the current subject believes that the candidate image features corresponding to the various image features to be configured are not compatible with the subject's real image features, the "image upload button" is clicked. The target software determines whether the current subject has the authorization to collect images. If not, an authorization pop-up window is displayed for the current subject to authorize; if authorized, the real image image corresponding to the current subject is collected, and the current image features corresponding to the current subject are determined based on the real image image.

[0040] In another embodiment, the current image feature configuration page includes a random image button. If the current object does not want to perform the image feature configuration operation, the "random image button" is clicked. The target software will randomly combine the candidate image features to determine the current image feature corresponding to the current object.

[0041] In another embodiment, the current object logs in to the target software and implements an image feature configuration request on the target software. The target software generates a current image feature configuration page corresponding to the current object according to the image feature configuration request. The current image feature configuration page displays an image feature customization area. The current object performs a custom editing operation in the image feature customization area according to its own real image features. The target software determines the current image feature corresponding to the current object according to the custom editing operation of the current object in the image feature customization area. For example, the current object has customized and edited "youth, golden middle-parted big waves, athletic system" in the image feature customization area. The target software accordingly determines the current image features corresponding to the current object as "youth", "golden middle-parted big waves" and "athletic system".

[0042] Determine other image features corresponding to other objects other than the current object, including but not limited to the following implementations:

[0043] In one embodiment, the current object implements a group photo invitation request on the target software. The target software displays a list of objects that can be invited according to the group photo invitation request. The current object selects at least one object from the object list as the other object to be invited according to the demand. The target software sends a target page link to the client corresponding to each other object. If the other object wishes to take a virtual image group photo with the current object, the target software implements a selection operation on the target page link. According to the selection operation of the other object on the target page link, the target software generates other image feature configuration pages corresponding to the other object.

[0044] The other image feature configuration page displays at least one type of image feature to be configured, and other objects perform configuration operations on each type of image feature to be configured according to their own real image features. The target software determines other image features corresponding to other objects according to the configuration operations performed by other objects on each type of image feature to be configured.

[0045] For optional implementation methods of determining other image features, reference may be made to the optional implementation methods of determining the current image features, which will not be described in detail here.

[0046] It is worth noting that in this embodiment, the image features of the current object and other objects are determined based on the image features authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of the image features comply with the relevant laws, regulations and standards of the relevant countries and regions, take necessary confidentiality measures, do not violate public order and good morals, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0047] S102: Generate a current virtual image prototype corresponding to the current object according to the current image features, and generate other virtual image prototypes corresponding to other objects according to other image features.

[0048] The virtual image prototype refers to the three-dimensional image prototype corresponding to the virtual image created by digital technology. It can be understood that the current virtual image prototype refers to the three-dimensional image prototype created by digital technology based on the current image characteristics; other virtual image prototypes refer to the three-dimensional image prototypes created by digital technology based on other image characteristics.

[0049] In one embodiment, the target software uses the API (Application Programming Interface) corresponding to the target image generation model to call the target image generation model, and then inputs the current image features as description text into the target image generation model. The target image generation model outputs a predicted virtual image prototype based on the description text as the current virtual image prototype corresponding to the current object; and inputs other image features as description text into the target image generation model. The target image generation model outputs predicted virtual image prototypes based on the description text as other virtual image prototypes corresponding to other objects.

[0050] Among them, large image generation models refer to deep learning models with a large parameter scale, which are mainly used to generate high-quality images from descriptive texts. They usually adopt a Transformer-based architecture, which can efficiently process and understand complex language descriptions and convert them into visual information. Usually, the parameter scale of large image generation models can even reach hundreds of billions.

[0051] Optionally, the target software also includes a "my image preview interface", and after the current virtual image prototype is generated, it is displayed in the "my image preview interface". The current object can view the current virtual image prototype in the "my image preview interface".

[0052] Optionally, during the process of generating the current virtual image prototype from the target image generating large model, the target software displays a loading status prompt in the "My Image Preview Interface" to prompt the waiting time required for the current virtual image prototype to be generated, wherein the waiting time can be estimated based on the performance of the target image generating large model, and can also be set based on experience, such as 15-20 seconds.

[0053] Optionally, during the process of generating the current virtual image prototype from the target image generation model, if the target software recognizes that the current object has implemented an image feature configuration request, an unavailable prompt message will be displayed to prompt that the current object cannot perform image feature configuration at this time, such as "My image is being generated, please try again later".

[0054] Optionally, during the process of generating the current virtual image prototype from the target image generation model, if the target software recognizes that the current object has implemented a photo invitation request, an unavailable prompt message will be displayed to prompt the current object that it is unable to invite other objects to take a virtual image photo at this time, such as "My image is being generated, please try again later".

[0055] S103: Generate a virtual image photo of the current object and other objects based on the current virtual image prototype and other virtual image prototypes.

[0056] In one embodiment, the number of subjects participating in the photo with the virtual image is determined, wherein the number of subjects is determined according to the number of the current subject and the number of other subjects. For example, if the number of other subjects is 3, the number of subjects is 3+1=4; if the number of other subjects is 2, the number of subjects is 2+1=3.

[0057] According to the current image features and other image features, the prototype size ratio between the current virtual image prototype and other virtual image prototypes is determined to ensure the degree of simulation of the virtual image photo to the real world. For example, assuming that the age feature of the current object is a teenager, and the age feature of any other object is a young adult, the prototype size of the current virtual image prototype of the current object should be smaller than the prototype size of the other virtual image prototypes of the other object, and then the prototype size ratio is estimated. For another example, assuming that the gender feature of other object A is female, and the gender feature of other object B is male, the prototype size of other virtual image prototypes of other object A should be smaller than the prototype size of other virtual image prototypes of other object B, and then the prototype size ratio is estimated.

[0058] Furthermore, according to the size of the group photo, the position of the group photo, the number of group photo objects, and the size ratio of the prototypes, the regional coordinates of the current virtual image prototype and each other virtual image prototype in the group photo are determined, and the regional coordinates are expressed in the form of x, y and z three-dimensional coordinates. According to the regional coordinates corresponding to the current virtual image prototype and each other virtual image prototype, a current group photo layer containing the current virtual image prototype and other group photo layers containing each other virtual image prototype are generated respectively, and then the current group photo layer and the other group photo layers are layered to generate a virtual image group photo between the current object and each other object.

[0059] Optionally, the target software also includes a "My Group Photo Interface", after generating a virtual image group photo between the current object and each other object, it is displayed in the "My Group Photo Interface". The current object can view the virtual image group photo in the "My Group Photo Interface".

[0060] Optionally, during the process of generating the virtual image group photo, the target software displays a loading status prompt in the "My Group Photo Interface" to prompt the waiting time required for the virtual image group photo to be generated, wherein the waiting time can be estimated in real time according to the progress of the group photo generation, or can be set according to experience. At the same time, the loading status prompt is also used to prompt that the final generated virtual image group photo can be viewed in "My Album". For example, "The virtual image group photo is being generated, estimated to take xx seconds, and the result can also be viewed later in [My Album]".

[0061] Optionally, the target software also includes a "My Album Icon", and "My Album" can be entered by clicking on the "My Album Icon". A bubble effect is added to the "My Album Icon" to guide the current object to click on the "My Album Icon" to enter "My Album" and view the result of the virtual image group photo generation. The bubble effect can be set to display for a time, such as 3 seconds.

[0062] Optionally, the current object clicks the "My Album Icon" to enter "My Album", and the target software determines whether the virtual image photo has been generated. If it has been generated, the virtual image photo will be displayed in the "My Album" interface; if it has not been generated, a generation waiting prompt message will be displayed in the "My Album" interface to prompt the waiting time required for the virtual image photo to be generated, such as "The virtual image photo is being generated, estimated to take xx seconds, please wait patiently", etc.

[0063] Optionally, if the virtual image group photo is generated, a "share button" will be displayed on the "my photo album" interface. If the current object clicks the "share button", the target software will share the virtual image group photo with the user to be shared selected by the current object.

[0064] Since the current virtual image prototype is generated based on the current image features corresponding to the current object, and other virtual image prototypes are generated based on other image features corresponding to other objects, the current virtual image prototype can realistically restore the real image features of the current object, and other virtual image prototypes can realistically restore the real image features of other objects, thereby achieving the effect of enhancing the humanization of virtual image group photos. Compared with the prior art of generating virtual image group photos through "AI face-changing", this embodiment makes the virtual image group photos closer to the real images of the objects participating in the photo, thereby improving the user experience.

[0065] Figure 2A It is a flowchart of other methods for taking group photos with virtual images disclosed in the embodiments of the present disclosure, which can be used to further optimize and expand the above technical solution and can be combined with the above-mentioned various optional implementation methods.

[0066] like Figure 2A As shown, the method for taking a group photo of a virtual image disclosed in this embodiment may include:

[0067] S201. Generate a current image feature configuration page corresponding to the current object according to an image feature configuration request implemented by the current object.

[0068] The current image feature configuration page includes at least one type of image feature to be configured.

[0069] In one embodiment, the target software includes a "customized exclusive image icon" in the "my image preview interface", and the current object clicks the "customized exclusive image icon" to implement the image feature configuration request. The target software generates a current image feature configuration page corresponding to the current object according to the image feature configuration request implemented by the current object.

[0070] S202: Determine the current image feature corresponding to the current object according to the configuration operation of the current object on each image feature to be configured.

[0071] In one implementation, the current object performs configuration operations on various image features to be configured in the current image feature configuration page, and the target software determines the current image feature corresponding to the current object according to the configuration operations of the current object on each image feature to be configured.

[0072] By generating a current image feature configuration page corresponding to the current object according to an image feature configuration request implemented by the current object; determining the current image feature corresponding to the current object according to the configuration operation of the current object on each image feature to be configured, thereby transferring the configuration right of the image feature to the current object, so that the current object can configure the image feature according to its own real image, and can also configure the image feature according to the image it expects to present in group photos, which greatly enhances the diversity and fun of the image feature configuration and further improves the user experience.

[0073] Optionally, the image features to be configured include at least one of gender features to be configured, age features to be configured, appearance features to be configured, and style features to be configured.

[0074] Among them, gender characteristics, age characteristics, appearance characteristics and style characteristics can all be configured independently by the current object.

[0075] By setting the image features to be configured to include at least one of the gender features to be configured, the age features to be configured, the appearance features to be configured and the style features to be configured, the types of configurable image features are enriched, the compatibility of the current virtual image prototype with the real image of the current object is ensured, and the humanization of the virtual image photo is further enhanced.

[0076] Optionally, each image feature to be configured includes at least one candidate image feature.

[0077] Optionally, the gender features to be configured include candidate image features "male" and "female". The age features to be configured include candidate image features "juvenile", "young" and "elderly". The appearance features to be configured include "middle-parted big waves", "long black straight hair", "round face wearing glasses", "short hair elf ears" and "thick eyebrows and big eyes". The style features to be configured include candidate image features "sporty", "simple style" and "lady style".

[0078] Optionally, according to the configuration operation of the current object on each image feature to be configured, determining the current image feature corresponding to the current object includes:

[0079] According to the selection operation of the current object on each candidate image feature, the current image feature corresponding to the current object is determined.

[0080] In one implementation, the current image feature configuration page displays each candidate image feature, the current object selects from each candidate image feature, and the target software determines the current image feature corresponding to the current object according to the selection operation of the current object.

[0081] For example, assuming that the current object has selected the candidate image features "female", "teenager", "round face wearing glasses" and "athletic", then the current image features corresponding to the current object are determined to be "female", "teenager", "round face wearing glasses" and "athletic".

[0082] By determining the current image feature corresponding to the current object based on the selection operation of each candidate image feature of the current object, the current object does not need to manually configure the image feature. The current image feature can be determined by selecting from the displayed candidate image features, which improves the efficiency of determining the current image feature and reduces the workload of configuring the image feature of the current object.

[0083] Optionally, according to the configuration operation of the current object on each image feature to be configured, determining the current image feature corresponding to the current object includes:

[0084] According to the user-defined editing operation of the current object on each image feature to be configured, the current image feature corresponding to the current object is determined.

[0085] In one embodiment, the current image feature configuration page includes an image feature customization area, and the current object performs a custom editing operation of the image feature in the image feature customization area according to its own real image features. The target software determines the current image feature corresponding to the current object according to the custom editing operation of the current object in the image feature customization area. For example, the current object customizes the image feature in the image feature customization area: "Youth, golden middle-parted big waves, athletic system", and the target software correspondingly determines the current image feature corresponding to the current object as "youth", "golden middle-parted big waves" and "athletic system".

[0086] By custom editing the image features to be configured according to the current object, the current image features corresponding to the current object are determined, so that users can customize the image features independently, which improves the flexibility of image feature configuration and further ensures the degree of compatibility between the current image features and the user's own real image.

[0087] Optionally, the method further includes:

[0088] According to the random feature generation request implemented by the current object, the candidate image features are randomly combined; and the current image feature corresponding to the current object is determined according to the random combination result.

[0089] In one embodiment, the current image feature configuration page includes a "random image button", and the current object clicks the "random image button" to implement a random feature generation request. The target software randomly combines each candidate image feature according to the implemented random feature generation request to obtain a set of random combination results of candidate image features, and uses the random combination results of the candidate image features as the current image feature corresponding to the current object.

[0090] For example, assuming that the candidate image features include "boy", "youth", "old", "middle-parted big waves", "long straight black hair", "round face wearing glasses", "short hair elf ears", "thick eyebrows and big eyes", "athletic", "simple style" and "lady style", if the current object clicks the "random image button", the target software will randomly combine the candidate image features. Assuming the random combination result is "boy", "short hair elf ears" and "simple style", "boy", "short hair elf ears" and "simple style" will be used as the current image features corresponding to the current object.

[0091] By generating a random feature request based on the current object, the candidate image features are randomly combined; the current image feature corresponding to the current object is determined based on the random combination result. Therefore, when the current object does not want to configure the image feature, the random image feature can be generated by randomly combining the candidate image features, so that the user does not need to configure the image feature independently, which speeds up the efficiency of determining the current image feature and meets the needs of the user.

[0092] Optionally, determining a current image feature corresponding to the current object includes:

[0093] Obtain a real image corresponding to the current object; and determine the current image features corresponding to the current object according to the real image.

[0094] In one embodiment, the current image feature configuration page includes an "image upload button", and the current subject clicks the "image upload button", and the target software determines whether the current subject has the authorization to collect images. If not, an authorization pop-up window is displayed for the current subject to authorize; if authorized, a selection panel "take a photo and upload" or "select from the album" is displayed, and the current subject uses the "take a photo and upload" panel or the "select from the album" panel to upload the real image. The target software inputs the real image image corresponding to the current subject into the pre-trained image feature extraction network, and uses the image feature extraction network to determine the current image feature corresponding to the current subject.

[0095] By acquiring the real image corresponding to the current object and determining the current image features corresponding to the current object according to the real image image, the technical means for determining the current image features are enriched and the flexibility of determining the current image features is improved.

[0096] S203: Generate a target page link according to the photo invitation request implemented by the current object.

[0097] Among them, the target page link is used to generate other image feature configuration pages corresponding to other objects.

[0098] In one implementation, the target software includes a "New Group Photo Button" in the "My Image Preview Interface", and the current object clicks the "New Group Photo Button" to implement the group photo invitation request. The target software displays a list of objects that can be invited according to the group photo invitation request, and the current object selects at least one object from the object list as the other object to be invited according to the requirements. The target software sends a target page link to the client corresponding to each other object, and the target page link can be a URL (Uniform Resource Locator) corresponding to other image feature configuration pages, etc.

[0099] S204: Generate other image feature configuration pages according to the selection operation performed by other objects on the target page link.

[0100] Among them, the other image feature configuration page contains at least one type of image feature to be configured.

[0101] In one embodiment, if other objects want to take a photo with the current object, they can select the target page link. The target software generates other image feature configuration pages corresponding to other objects according to the selection operation of the other objects on the target page link.

[0102] S205: Determine other image features corresponding to other objects according to the configuration operations of other objects on each image feature to be configured.

[0103] Among them, the optional implementation of S205 can refer to the optional implementation of determining the current image characteristics, which will not be repeated here.

[0104] By generating a target page link based on the photo invitation request implemented by the current object; generating other image feature configuration pages based on the selection operations implemented by other objects on the target page link; determining other image features corresponding to other objects based on the configuration operations of other objects on each image feature to be configured, thereby utilizing the photo invitation mechanism to determine other image features of other objects, thereby increasing interactivity between users and further improving user experience.

[0105] S206: Generate a current virtual image prototype corresponding to the current object according to the current image features, and generate other virtual image prototypes corresponding to other objects according to other image features.

[0106] S207: Generate a virtual image photo of the current object and other objects based on the current virtual image prototype and other virtual image prototypes.

[0107] Optionally, generating a virtual image photo of the current object and the other object based on the current virtual image prototype and the other virtual image prototypes, including:

[0108] According to a photo background selection request implemented by the current object, a target photo background is determined from candidate photo backgrounds; and according to the target photo background, the current virtual image prototype and other virtual image prototypes, a virtual image photo between the current object and other objects is generated.

[0109] Among them, the candidate photo background refers to the photo background that can be selected by the current object, and is used to switch the photo background of the virtual image photo.

[0110] In one embodiment, the target software displays background thumbnails of candidate group photo backgrounds in the "My Image Preview Interface". The current subject implements a group photo background selection request for any background thumbnail of the candidate group photo background according to preference, and the target software uses the candidate group photo background corresponding to the background thumbnail selected by the current subject as the target group photo background according to the group photo background selection request implemented by the current subject. Furthermore, the target software generates a virtual image group photo between the current subject and other subjects according to the target group photo background, the current virtual image prototype and other virtual image prototypes, so that the background of the virtual image group photo appears as the target group photo background.

[0111] By determining a target photo background from candidate photo backgrounds based on a photo background selection request implemented by the current object; generating a virtual image photo between the current object and other objects based on the target photo background, the current virtual image prototype and other virtual image prototypes, the current object can select a photo background for the virtual image photo according to his or her own preferences, thereby enhancing the fun and liveliness of the virtual image photo.

[0112] Optionally, after determining a target photo background from candidate photo backgrounds according to the photo background selection request implemented by the current object, the method further includes:

[0113] According to the correlation between the background of the group photo and the clothing features, the target clothing features associated with the background of the target group photo are determined.

[0114] The association between the background of the group photo and the clothing features is pre-established, that is, the clothing features used to generate the virtual image prototype can be directly determined according to any group photo background. For example, assuming that the background of any group photo is "New Year background", the clothing features associated with the group photo background can be "Tang suit", "festive clothing", etc.

[0115] By determining the target clothing features associated with the target photo background based on the correlation between the photo background and the clothing features, the target clothing features can be adapted to the target photo background, further improving the aesthetics of the virtual image photo.

[0116] Optionally, generating a current virtual image prototype corresponding to the current object according to the current image feature, and generating other virtual image prototypes corresponding to other objects according to other image features, includes:

[0117] According to the current image features and the target clothing features, a current virtual image prototype corresponding to the current object is generated; and according to other image features and the target clothing features, other virtual image prototypes corresponding to other objects are generated.

[0118] In one embodiment, the image generation model is used to predict the virtual image prototype based on the current image features and the target clothing features, so as to generate the current virtual image prototype corresponding to the current object; and, based on other image features and the target clothing features, the image generation model is used to predict the virtual image prototype, so as to generate other virtual image prototypes corresponding to other objects.

[0119] By generating the current virtual image prototype corresponding to the current object according to the current image features and the target clothing features; and generating other virtual image prototypes corresponding to other objects according to other image features and the target clothing features, the virtual image prototype is made more vivid and the aesthetics of the virtual image prototype is increased.

[0120] Optionally, generating a current virtual image prototype corresponding to the current object according to the current image features and the target clothing features, including:

[0121] A first description text is generated according to the current image features and the target clothing features, and the first description text is input into the image generation model; and a current virtual image prototype corresponding to the current object is determined according to the first predicted virtual image prototype output by the image generation model.

[0122] Optionally, based on other image features and target clothing features, other virtual image prototypes corresponding to other objects are generated, including:

[0123] A second description text is generated according to other image features and target clothing features, and the second description text is input into the image generation model; and other virtual image prototypes corresponding to other objects are determined according to the second predicted virtual image prototype output by the image generation model.

[0124] In one embodiment, the target software uses the API corresponding to the image generation model to call the image generation model, and then uses the current image features and the target clothing features as the first description text, which are input into the image generation model. The image generation model outputs a first predicted virtual image prototype based on the first description text, which is used as the current virtual image prototype corresponding to the current object; and uses other image features and the target clothing features as the second description text, which are input into the image generation model. The image generation model outputs a second predicted virtual image prototype based on the second description text, which are used as other virtual image prototypes corresponding to other objects.

[0125] Since the large image generation model has a huge parameter scale, even reaching the trillion level, and has good model performance, a first description text is generated according to the current image features and the target clothing features, and the first description text is input into the large image generation model; the current virtual image prototype corresponding to the current object is determined according to the first predicted virtual image prototype output by the large image generation model; and, according to other image features and the target clothing features, a second description text is generated, and the second description text is input into the large image generation model; according to the second predicted virtual image prototype output by the large image generation model, other virtual image prototypes corresponding to other objects are determined, which can improve the adaptability of the current virtual image prototype to the real image of the current object, as well as the adaptability of other virtual image prototypes to the real images of other objects, further improving the humanization of virtual image group photos.

[0126] Optionally, after determining a target photo background from candidate photo backgrounds according to the photo background selection request implemented by the current object, the method further includes:

[0127] The upper limit of the number of target photo objects associated with the target photo background is determined according to the association between the photo background and the upper limit of the number of photo objects.

[0128] The upper limit of the number of subjects in the group photo refers to the maximum number of subjects that can participate in the group photo of the virtual image. The association between the group photo background and the upper limit of the number of subjects in the group photo is established in advance, that is, the corresponding upper limit of the number of subjects in the group photo can be directly determined according to any group photo background. For example, if the background of any group photo is "outdoor background", a high upper limit of the number of subjects in the group photo can be set; for another example, if the background of any group photo is "indoor background", a low upper limit of the number of subjects in the group photo can be set, etc.

[0129] By determining the upper limit of the target photo objects associated with the target photo background based on the correlation between the photo background and the upper limit of the number of photo objects, the upper limit of the target photo objects can be adapted to the target photo background, further improving the aesthetics of the virtual image photo.

[0130] Optionally, based on the photo invitation request implemented by the current object, a target page link is generated, including:

[0131] If the number of photo objects in the virtual image photo is less than the upper limit of the number of photo objects, a target page link is generated according to the photo invitation request implemented by the current object.

[0132] In one embodiment, the target software determines the relationship between the number of photo objects in the virtual image group photo and the upper limit of the number of photo objects. If it is determined that the number of photo objects in the virtual image group photo is equal to the upper limit of the number of photo objects, the photo invitation request implemented by the current object is not supported; if it is determined that the number of photo objects in the virtual image group photo is equal to the upper limit of the number of photo objects, then according to the photo invitation request implemented by the current object, a target page link is generated to continue to add new other objects in the virtual image group photo.

[0133] By generating a target page link based on the photo invitation request implemented by the current object when the number of photo objects in the virtual image group photo is less than the upper limit of the number of photo objects, the problem of unlimited photo invitations resulting in too many objects in the virtual image group photo is avoided, thereby further ensuring the aesthetics of the virtual image group photo.

[0134] Figure 2B is a schematic diagram of some current image feature configuration pages disclosed according to the embodiments of the present disclosure, such as Figure 2B As shown, the current image feature configuration page 200 includes a gender feature 201 to be configured, an age feature 202 to be configured, an appearance feature 203 to be configured, and a style feature 204 to be configured.

[0135] The current image feature configuration page 200 also includes candidate image features 205 corresponding to the gender feature 201 to be configured, candidate image features 206 corresponding to the age feature 202 to be configured, candidate image features 207 corresponding to the appearance feature 203 to be configured, and candidate image features 208 corresponding to the style feature 204 to be configured.

[0136] The current avatar configuration page 200 also includes an avatar customization area 209 .

[0137] The current avatar feature configuration page 200 also includes a random avatar button 210 .

[0138] The current avatar configuration page 200 also includes an image upload button 211 .

[0139] Figure 3 It is a flowchart of other methods for taking group photos with virtual images disclosed in the embodiments of the present disclosure, which can be used to further optimize and expand the above technical solution and can be combined with the above-mentioned various optional implementation methods.

[0140] like Figure 3 As shown, the method for taking a group photo of a virtual image disclosed in this embodiment may include:

[0141] S301: Determine a current image feature corresponding to a current object, and other image features corresponding to other objects except the current object.

[0142] S302, determining the current photo position of the current object in the virtual image photo, and other photo positions of other objects in the virtual image photo; and determining the current action feature associated with the current photo position, and other action features associated with other photo positions according to the association relationship between the photo position and the action feature.

[0143] Among them, the photo position is used to describe the position of the object in the virtual image photo, such as the center position, the leftmost position, the left or right position, the second position from the left, etc. The photo position can be set by the current object according to preference, or it can be determined according to the order in which the virtual image prototype is generated, etc. For example, the photo position of the object that generates the virtual image prototype first is closer to the center position. It can be understood that the current photo position is used to describe the position of the current object in the virtual image photo; other photo positions are used to describe the positions of other objects in the virtual image photo.

[0144] The association between the photo position and the action characteristics is established in advance, that is, the action characteristics used to generate the virtual image prototype can be directly determined according to any photo position. For example, the action characteristics of the virtual image prototype located in the center of the photo are usually looking straight ahead, while the action characteristics of the virtual image prototype near the left or right side of the photo are usually looking sideways to ensure the simulation of the virtual image photo.

[0145] In one implementation, the target software determines the action feature associated with the current group photo position as the current action feature, and determines the action features associated with other group photo positions as other action features, based on the association relationship between the group photo positions and the action features.

[0146] S303: Generate a current virtual image prototype corresponding to the current object according to the current image features and the current action features, and generate other virtual image prototypes corresponding to other objects according to other image features and other action features.

[0147] In one embodiment, an image generation model is used to predict a virtual image prototype based on current image features and current action features, so as to generate a current virtual image prototype corresponding to the current object; and an image generation model is used to predict a virtual image prototype based on other image features and other action features, so as to generate other virtual image prototypes corresponding to other objects.

[0148] By determining the current photo position of the current object in the virtual image photo, and other photo positions of other objects in the virtual image photo; according to the correlation relationship between the photo position and the action feature, determining the current action feature associated with the current photo position, and other action features associated with other photo positions; generating the current virtual image prototype corresponding to the current object according to the current image feature and the current action feature, and generating other virtual image prototypes corresponding to other objects according to other image features and other action features, the virtual image prototype is made more vivid and the aesthetics of the virtual image prototype is increased.

[0149] Optionally, generating a current virtual image prototype corresponding to the current object according to the current image feature and the current action feature includes:

[0150] A third description text is generated according to the current image features and the current action features, and the third description text is input into the image generation model; based on the third predicted virtual image prototype output by the image generation model, the current virtual image prototype corresponding to the current object is determined.

[0151] Optionally, based on other image features and other action features, other virtual image prototypes corresponding to other objects are generated, including:

[0152] A fourth description text is generated according to other image features and other action features, and the fourth description text is input into the image generation model; and other virtual image prototypes corresponding to other objects are determined according to the fourth predicted virtual image prototype output by the image generation model.

[0153] In one embodiment, the target software uses the API corresponding to the image generation model to call the image generation model, and then inputs the current image features and the current action features as a third description text into the image generation model. The image generation model outputs a third predicted virtual image prototype according to the third description text as the current virtual image prototype corresponding to the current object; and inputs other image features and other action features as a fourth description text into the image generation model. The image generation model outputs a fourth predicted virtual image prototype according to the fourth description text as other virtual image prototypes corresponding to other objects.

[0154] Since the large image generation model has a huge parameter scale, even reaching the trillion level, and has good model performance, a third description text is generated according to the current image features and the current action features, and the third description text is input into the large image generation model; the current virtual image prototype corresponding to the current object is determined according to the third predicted virtual image prototype output by the large image generation model, and a fourth description text is generated according to other image features and other action features, and the fourth description text is input into the large image generation model; other virtual image prototypes corresponding to other objects are determined according to the fourth predicted virtual image prototype output by the large image generation model, which can improve the adaptability of the current virtual image prototype to the real image of the current object, as well as the adaptability of other virtual image prototypes to the real images of other objects, further improving the humanization of virtual image group photos.

[0155] S304: Determine the number of subjects to be photographed, and the prototype size ratio between the current avatar prototype and other avatar prototypes.

[0156] Among them, the number of subjects in the photo refers to the number of subjects participating in the photo with the virtual image.

[0157] In one embodiment, the prototype size ratio between the current virtual image prototype and other virtual image prototypes is determined based on the current image features and other image features to ensure the degree of simulation of the virtual image group photo to the real world. The prototype size ratio can be represented by the coordinate ratio of the virtual image prototype in the x, y and z coordinate directions.

[0158] For example, assuming that the age feature of the current object is a teenager, and the age feature of any other object is a young adult, the prototype size of the current virtual image prototype of the current object should be smaller than the prototype size of the other virtual image prototypes of the other object, and then the prototype size ratio is estimated. For another example, assuming that the gender feature of other object A is female, and the gender feature of other object B is male, the prototype size of the other virtual image prototype of other object A should be smaller than the prototype size of the other virtual image prototype of other object B, and then the prototype size ratio is estimated.

[0159] S305. Determine the current area coordinates of the current virtual image prototype in the group photo based on the size of the group photo, the number of group photo objects, the prototype size ratio and the current group photo position; determine the other area coordinates of other virtual image prototypes in the group photo based on the size of the group photo, the number of group photo objects, the prototype size ratio and other group photo positions.

[0160] The photo size refers to the size of the final generated virtual image photo.

[0161] In one embodiment, according to the size of the group photo, the number of objects in the group photo, the size ratio of the prototype and the current position in the group photo, the preset group photo image processing logic is used to determine the current area coordinates of the current virtual image prototype in the group photo, and the area coordinates are expressed in the form of x, y and z three-dimensional coordinates.

[0162] Furthermore, based on the size of the group photo, the number of people in the photo, the size ratio of the prototypes and other positions in the group photo, the preset group photo image processing logic is used to determine the coordinates of other regions where other virtual image prototypes are located in the group photo, and the region coordinates are expressed in the form of x, y and z three-dimensional coordinates.

[0163] S306: Generate a virtual image photo of the current object and other objects based on the current region coordinates and the current virtual image prototype, as well as other region coordinates and other virtual image prototypes.

[0164] In one embodiment, the current virtual image prototype and the current region coordinates are input into the layer generation model, and a layer containing the current virtual image prototype is generated as the current group photo layer; and other virtual image prototypes and other region coordinates are input into the layer generation model, and layers containing other virtual image prototypes are generated as other group photo layers. Furthermore, the current group photo layer and other group photo layers are layer-overlaid to generate virtual image group photos between the current object and other objects.

[0165] By determining the number of group photo objects, and the prototype size ratio between the current virtual image prototype and other virtual image prototypes; determining the current regional coordinates of the current virtual image prototype in the group photo according to the group photo size, the number of group photo objects, the prototype size ratio and the current group photo position; determining other regional coordinates of other virtual image prototypes in the group photo according to the group photo size, the number of group photo objects, the prototype size ratio and other group photo positions; generating a virtual image group photo between the current object and the other objects according to the current regional coordinates and the current virtual image prototype, as well as other regional coordinates and other virtual image prototypes, thereby ensuring that each virtual image prototype is in a suitable regional position in the virtual image group photo, and further ensuring the aesthetics of the virtual image group photo.

[0166] Optionally, the method further includes:

[0167] A. According to a photo download request implemented by the current object, a target virtual image photo is determined from at least one historical virtual image photo.

[0168] Among them, each historical virtual image group photo is generated according to the current virtual image prototype and a different number of other virtual image prototypes. For example, suppose that there are historical virtual image group photo 1, historical virtual image group photo 2, historical virtual image group photo 3 and historical virtual image group photo 4, among which historical virtual image group photo 1 is generated according to the current virtual image prototype and 1 other virtual image prototype; historical virtual image group photo 2 is generated according to the current virtual image prototype and 2 other virtual image prototypes; historical virtual image group photo 3 is generated according to the current virtual image prototype and 3 other virtual image prototypes; historical virtual image group photo 4 is generated according to the current virtual image prototype and 4 other virtual image prototypes.

[0169] In one embodiment, the current object implements a photo download request for at least one historical virtual image photo that the current object wants to download in "My Album", and the target software uses the historical virtual image photo for which the current object implements the photo download request as the target virtual image photo based on the photo download request implemented by the current object.

[0170] B. Perform a photo download operation on the photo taken with the target virtual image.

[0171] In one implementation, a group photo downloading operation is performed by taking a group photo of the target virtual image, so as to download the group photo of the target virtual image to a target terminal set by the current object.

[0172] By determining a target virtual image group photo from at least one historical virtual image group photo according to a group photo download request implemented by the current object, and performing a group photo download operation on the target virtual image group photo, the user's customized group photo download needs are met, so that the user can download virtual image group photos with different numbers of photo objects as needed.

[0173] Figure 4 The present invention is a schematic diagram of a virtual image group photo device according to an embodiment of the present invention, which can be applied to generate a virtual image group photo between a current object and other objects. The device of this embodiment can be implemented by software and / or hardware, and can be integrated on any electronic device with computing capabilities.

[0174] like Figure 4 As shown, the virtual image group photo device 40 disclosed in this embodiment may include an image feature determination module 41, a virtual image prototype generation module 42 and a virtual image group photo generation module 43, wherein:

[0175] An image feature determination module 41 is used to determine a current image feature corresponding to a current object, and other image features corresponding to other objects except the current object;

[0176] A virtual image prototype generating module 42 is used to generate a current virtual image prototype corresponding to the current object according to the current image feature, and to generate other virtual image prototypes corresponding to the other objects according to the other image features;

[0177] The virtual image group photo generating module 43 is used to generate a virtual image group photo between the current object and the other objects according to the current virtual image prototype and the other virtual image prototypes.

[0178] Optionally, the image feature determination module 41 is specifically used for:

[0179] According to the image feature configuration request implemented by the current object, a current image feature configuration page corresponding to the current object is generated; wherein the current image feature configuration page includes at least one type of image feature to be configured;

[0180] According to the configuration operation of the current object on each of the to-be-configured image features, the current image feature corresponding to the current object is determined.

[0181] Optionally, each of the image features to be configured includes at least one candidate image feature;

[0182] The image feature determination module 41 is further specifically used for:

[0183] According to the selection operation of the current object on each of the candidate image features, the current image feature corresponding to the current object is determined.

[0184] Optionally, the image feature determination module 41 is further configured to:

[0185] According to the user-defined editing operation of the current object on each of the image features to be configured, the current image feature corresponding to the current object is determined.

[0186] Optionally, the device further includes a feature random generation module, which is specifically used to:

[0187] Randomly combining the candidate image features according to the random feature generation request implemented by the current object;

[0188] The current image feature corresponding to the current object is determined according to the random combination result.

[0189] Optionally, the image feature determination module 41 is further configured to:

[0190] Obtaining a real image corresponding to the current object;

[0191] The current image feature corresponding to the current object is determined according to the real image.

[0192] Optionally, the image feature determination module 41 is further configured to:

[0193] Generate a target page link according to the photo invitation request implemented by the current object; wherein the target page link is used to generate other image feature configuration pages corresponding to the other objects;

[0194] According to the selection operation performed by the other object on the target page link, the other image feature configuration page is generated; wherein the other image feature configuration page contains at least one type of image feature to be configured;

[0195] According to the configuration operation of the other objects on each of the image features to be configured, other image features corresponding to the other objects are determined.

[0196] Optionally, the virtual image group photo generating module 43 is specifically used for:

[0197] Determining a target group photo background from candidate group photo backgrounds according to a group photo background selection request implemented by the current object;

[0198] A virtual image group photo of the current object and the other objects is generated according to the target group photo background, the current virtual image prototype and the other virtual image prototypes.

[0199] Optionally, the device further comprises a clothing feature determination module, which is specifically used to:

[0200] Determining target clothing features associated with the target group photo background according to the association relationship between the group photo background and the clothing features;

[0201] Optionally, the virtual image prototype generation module 42 is specifically used for:

[0202] Generating a current virtual image prototype corresponding to the current object according to the current image features and the target clothing features;

[0203] And, based on the other image features and the target clothing features, other virtual image prototypes corresponding to the other objects are generated.

[0204] Optionally, the device further includes a quantity upper limit determination module, which is specifically configured to:

[0205] Determining the upper limit of the number of target group photo objects associated with the target group photo background according to the association relationship between the group photo background and the upper limit of the number of group photo objects;

[0206] Optionally, the image feature determination module 41 is further configured to:

[0207] If the number of the group photo objects in the virtual image group photo is less than the upper limit of the number of group photo objects, the target page link is generated according to the group photo invitation request implemented by the current object.

[0208] Optionally, the virtual image prototype generation module 42 is further configured to:

[0209] Determine a current photo position of the current object in the photo with the virtual image, and other photo positions of the other objects in the photo with the virtual image;

[0210] According to the association relationship between the group photo position and the action feature, determine the current action feature associated with the current group photo position, and other action features associated with the other group photo positions;

[0211] A current virtual image prototype corresponding to the current object is generated according to the current image feature and the current action feature, and other virtual image prototypes corresponding to the other objects are generated according to the other image features and the other action features.

[0212] Optionally, the virtual image prototype generation module 42 is further configured to:

[0213] Generate a first description text according to the current image feature and the target clothing feature, and input the first description text into the image generation model;

[0214] Determine a current virtual image prototype corresponding to the current object according to a first predicted virtual image prototype output by the image generation large model;

[0215] Generate a second description text according to the other image features and the target clothing features, and input the second description text into the image generation model;

[0216] The second predicted virtual image prototype output by the image generation model is used to determine other virtual image prototypes corresponding to the other objects.

[0217] Optionally, the virtual image prototype generation module 42 is further configured to:

[0218] Generate a third description text according to the current image feature and the current action feature, and input the third description text into the image generation model;

[0219] Determine a current virtual image prototype corresponding to the current object according to a third predicted virtual image prototype output by the image generation large model;

[0220] Generate a fourth description text according to the other image features and the other action features, and input the fourth description text into the image generation model;

[0221] The other virtual image prototypes corresponding to the other objects are determined according to the fourth predicted virtual image prototype output by the image generation model.

[0222] Optionally, the virtual image group photo generating module 43 is further configured to:

[0223] Determine the number of subjects to be photographed, and the prototype size ratio between the current avatar prototype and the other avatar prototypes;

[0224] Determine the current region coordinates of the current avatar prototype in the group photo according to the size of the group photo, the number of subjects in the group photo, the size ratio of the prototype and the current position in the group photo;

[0225] Determine the coordinates of other regions where the other avatar prototypes are located in the group photo according to the size of the group photo, the number of the group photo subjects, the size ratio of the prototypes and the positions of the other group photos;

[0226] A virtual image photo of the current object and the other objects is generated according to the current region coordinates and the current virtual image prototype, and the other region coordinates and the other virtual image prototypes.

[0227] Optionally, the device further includes a group photo downloading module, which is specifically used to:

[0228] Determining a target virtual image group photo from at least one historical virtual image group photo according to a group photo download request implemented by the current object; wherein each of the historical virtual image group photos is generated according to the current virtual image prototype and a different number of other virtual image prototypes;

[0229] The group photo downloading operation is performed on the group photo of the target virtual image.

[0230] Optionally, the image features to be configured include at least one of gender features to be configured, age features to be configured, appearance features to be configured, and style features to be configured.

[0231] The virtual image group photo device 40 disclosed in the embodiment of the present disclosure can execute the virtual image group photo method disclosed in the embodiment of the present disclosure, and has the corresponding functional modules and beneficial effects of the execution method. For the contents not described in detail in this embodiment, reference can be made to the description in the embodiment of the method disclosed in the present disclosure.

[0232] In the technical solution disclosed herein, the acquisition, storage and application of user personal information involved are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0233] According to an embodiment of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium and a computer program product.

[0234] Figure 5 A schematic block diagram of an example electronic device 500 that can be used to implement an embodiment of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or required herein.

[0235] like Figure 5 As shown, the device 500 includes a computing unit 501, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 502 or a computer program loaded from a storage unit 508 into a random access memory (RAM) 503. In the RAM 503, various programs and data required for the operation of the device 500 can also be stored. The computing unit 501, the ROM 502, and the RAM 503 are connected to each other via a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0236] A number of components in the device 500 are connected to the I / O interface 505, including: an input unit 506, such as a keyboard, a mouse, etc.; an output unit 507, such as various types of displays, speakers, etc.; a storage unit 508, such as a disk, an optical disk, etc.; and a communication unit 509, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 509 allows the device 500 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0237] The computing unit 501 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 501 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 501 performs the various methods and processes described above, such as the virtual image photo method. For example, in some embodiments, the virtual image photo method may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit 508. In some embodiments, part or all of the computer program may be loaded and / or installed on the device 500 via the ROM 502 and / or the communication unit 509. When the computer program is loaded into the RAM 503 and executed by the computing unit 501, one or more steps of the virtual image photo method described above may be performed. Alternatively, in other embodiments, the computing unit 501 may be configured to perform the virtual image photo method in any other appropriate manner (e.g., by means of firmware).

[0238] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0239] The program code for implementing the method of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0240] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0241] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0242] The systems and techniques described herein may be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0243] A computer system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The relationship of client and server is generated by computer programs running on respective computers and having a client-server relationship with each other. The server may be a cloud server, a server of a distributed system, or a server combined with a blockchain.

[0244] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps recorded in this disclosure can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved, and this document does not limit this.

[0245] The above specific implementations do not constitute a limitation on the protection scope of the present disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A method for taking a group photo of a virtual image, comprising: Determine a current image feature corresponding to a current object, and other image features corresponding to other objects except the current object; Generating a current virtual image prototype corresponding to the current object according to the current image feature, and generating other virtual image prototypes corresponding to the other objects according to the other image features; A virtual image photo of the current object and the other objects is generated according to the current virtual image prototype and the other virtual image prototypes.

2. The method according to claim 1, wherein: The step of determining the current image feature corresponding to the current object includes: According to the image feature configuration request implemented by the current object, a current image feature configuration page corresponding to the current object is generated; wherein the current image feature configuration page includes at least one type of image feature to be configured; According to the configuration operation of the current object on each of the to-be-configured image features, the current image feature corresponding to the current object is determined.

3. The method according to claim 2, wherein: Each of the image features to be configured includes at least one candidate image feature; The step of determining the current image feature corresponding to the current object according to the configuration operation of the current object on each of the image features to be configured includes: According to the selection operation of the current object on each of the candidate image features, the current image feature corresponding to the current object is determined.

4. The method according to claim 2, wherein: The step of determining the current image feature corresponding to the current object according to the configuration operation of the current object on each of the image features to be configured includes: According to the user-defined editing operation of the current object on each of the image features to be configured, the current image feature corresponding to the current object is determined.

5. The method according to claim 3, further comprising: Randomly combining the candidate image features according to the random feature generation request implemented by the current object; The current image feature corresponding to the current object is determined according to the random combination result.

6. The method according to claim 1, wherein: The step of determining the current image feature corresponding to the current object includes: Obtaining a real image corresponding to the current object; The current image feature corresponding to the current object is determined according to the real image.

7. The method according to claim 1, wherein: The determining of other image features corresponding to other objects other than the current object includes: Generate a target page link according to the photo invitation request implemented by the current object; wherein the target page link is used to generate other image feature configuration pages corresponding to the other objects; According to the selection operation performed by the other object on the target page link, the other image feature configuration page is generated; wherein the other image feature configuration page contains at least one type of image feature to be configured; According to the configuration operation of the other objects on each of the image features to be configured, other image features corresponding to the other objects are determined.

8. The method according to claim 7, wherein: The step of generating a virtual image photo of the current object and the other object according to the current virtual image prototype and the other virtual image prototypes includes: Determining a target group photo background from candidate group photo backgrounds according to a group photo background selection request implemented by the current object; A virtual image group photo of the current object and the other objects is generated according to the target group photo background, the current virtual image prototype and the other virtual image prototypes.

9. The method according to claim 8, after determining a target group photo background from candidate group photo backgrounds according to the group photo background selection request implemented by the current object, further comprising: Determining target clothing features associated with the target group photo background according to the association relationship between the group photo background and the clothing features; The step of generating a current virtual image prototype corresponding to the current object according to the current image feature, and generating other virtual image prototypes corresponding to other objects according to the other image features, comprises: Generating a current virtual image prototype corresponding to the current object according to the current image features and the target clothing features; And, based on the other image features and the target clothing features, other virtual image prototypes corresponding to the other objects are generated.

10. The method according to claim 8, after determining a target group photo background from candidate group photo backgrounds according to the group photo background selection request implemented by the current object, further comprising: Determining the upper limit of the number of target group photo objects associated with the target group photo background according to the association relationship between the group photo background and the upper limit of the number of group photo objects; The step of generating a target page link according to the photo invitation request implemented by the current object includes: If the number of the group photo objects in the virtual image group photo is less than the upper limit of the number of group photo objects, the target page link is generated according to the group photo invitation request implemented by the current object.

11. The method according to claim 9, wherein: The step of generating a current virtual image prototype corresponding to the current object according to the current image feature, and generating other virtual image prototypes corresponding to other objects according to the other image features, comprises: Determine a current photo position of the current object in the photo with the virtual image, and other photo positions of the other objects in the photo with the virtual image; According to the association relationship between the group photo position and the action feature, determine the current action feature associated with the current group photo position, and other action features associated with the other group photo positions; A current virtual image prototype corresponding to the current object is generated according to the current image feature and the current action feature, and other virtual image prototypes corresponding to the other objects are generated according to the other image features and the other action features.

12. The method according to claim 9, wherein: The step of generating a current virtual image prototype corresponding to the current object according to the current image feature and the target clothing feature comprises: Generate a first description text according to the current image feature and the target clothing feature, and input the first description text into the image generation model; Determine a current virtual image prototype corresponding to the current object according to a first predicted virtual image prototype output by the image generation large model; The step of generating other virtual image prototypes corresponding to the other objects according to the other image features and the target clothing features includes: Generate a second description text according to the other image features and the target clothing features, and input the second description text into the image generation model; The second predicted virtual image prototype output by the image generation model is used to determine other virtual image prototypes corresponding to the other objects.

13. The method according to claim 11, wherein: The step of generating a current virtual image prototype corresponding to the current object according to the current image feature and the current action feature comprises: Generate a third description text according to the current image feature and the current action feature, and input the third description text into the image generation model; Determine a current virtual image prototype corresponding to the current object according to a third predicted virtual image prototype output by the image generation large model; The step of generating other virtual image prototypes corresponding to the other objects according to the other image features and the other action features includes: Generate a fourth description text according to the other image features and the other action features, and input the fourth description text into the image generation model; The other virtual image prototypes corresponding to the other objects are determined according to the fourth predicted virtual image prototype output by the image generation model.

14. The method according to claim 11, wherein: The step of generating a virtual image photo of the current object and the other object according to the current virtual image prototype and the other virtual image prototypes includes: Determine the number of subjects to be photographed, and the prototype size ratio between the current avatar prototype and the other avatar prototypes; Determine the current region coordinates of the current avatar prototype in the group photo according to the size of the group photo, the number of subjects in the group photo, the size ratio of the prototype and the current position in the group photo; Determine the coordinates of other regions where the other avatar prototypes are located in the group photo according to the size of the group photo, the number of the group photo subjects, the size ratio of the prototypes and the positions of the other group photos; A virtual image photo of the current object and the other objects is generated according to the current region coordinates and the current virtual image prototype, and the other region coordinates and the other virtual image prototypes.

15. The method according to claim 1, further comprising: Determining a target virtual image group photo from at least one historical virtual image group photo according to a group photo download request implemented by the current object; wherein each of the historical virtual image group photos is generated according to the current virtual image prototype and a different number of other virtual image prototypes; The group photo downloading operation is performed on the group photo of the target virtual image.

16. The method according to claim 2 or 7, wherein: The image features to be configured include at least one of gender features to be configured, age features to be configured, appearance features to be configured, and style features to be configured.

17. A virtual image group photo device, comprising: An image feature determination module, used to determine a current image feature corresponding to a current object, and other image features corresponding to other objects except the current object; A virtual image prototype generation module, used to generate a current virtual image prototype corresponding to the current object according to the current image feature, and to generate other virtual image prototypes corresponding to the other objects according to the other image features; The virtual image group photo generation module is used to generate a virtual image group photo between the current object and the other objects according to the current virtual image prototype and the other virtual image prototypes.

18. The device according to claim 17, wherein: The image feature determination module is specifically used for: According to the image feature configuration request implemented by the current object, a current image feature configuration page corresponding to the current object is generated; wherein the current image feature configuration page includes at least one type of image feature to be configured; According to the configuration operation of the current object on each of the to-be-configured image features, the current image feature corresponding to the current object is determined.

19. The device according to claim 18, wherein: Each of the image features to be configured includes at least one candidate image feature; The image feature determination module is further specifically used for: According to the selection operation of the current object on each of the candidate image features, the current image feature corresponding to the current object is determined.

20. The device according to claim 18, wherein The image feature determination module is further specifically used for: According to the user-defined editing operation of the current object on each of the image features to be configured, the current image feature corresponding to the current object is determined.

21. The device according to claim 19, further comprising a feature random generation module, specifically configured to: Randomly combining the candidate image features according to the random feature generation request implemented by the current object; The current image feature corresponding to the current object is determined according to the random combination result.

22. The device according to claim 17, wherein: The image feature determination module is further specifically used for: Obtaining a real image corresponding to the current object; The current image feature corresponding to the current object is determined according to the real image.

23. The device according to claim 17, wherein: The image feature determination module is further specifically used for: Generate a target page link according to the photo invitation request implemented by the current object; wherein the target page link is used to generate other image feature configuration pages corresponding to the other objects; According to the selection operation performed by the other object on the target page link, the other image feature configuration page is generated; wherein the other image feature configuration page contains at least one type of image feature to be configured; According to the configuration operation of the other objects on each of the image features to be configured, other image features corresponding to the other objects are determined.

24. The device according to claim 23, wherein: The virtual image group photo generation module is specifically used for: Determining a target group photo background from candidate group photo backgrounds according to a group photo background selection request implemented by the current object; A virtual image group photo of the current object and the other objects is generated according to the target group photo background, the current virtual image prototype and the other virtual image prototypes.

25. The device according to claim 24, further comprising a clothing feature determination module, specifically configured to: Determining target clothing features associated with the target group photo background according to the association relationship between the group photo background and the clothing features; The virtual image prototype generation module is specifically used for: Generating a current virtual image prototype corresponding to the current object according to the current image features and the target clothing features; And, based on the other image features and the target clothing features, other virtual image prototypes corresponding to the other objects are generated.

26. The device according to claim 24, further comprising a quantity upper limit determination module, specifically configured to: Determining the upper limit of the number of target group photo objects associated with the target group photo background according to the association relationship between the group photo background and the upper limit of the number of group photo objects; The image feature determination module is further specifically used for: If the number of the group photo objects in the virtual image group photo is less than the upper limit of the number of group photo objects, the target page link is generated according to the group photo invitation request implemented by the current object.

27. The device according to claim 25, wherein: The virtual image prototype generation module is further specifically used for: Determine a current photo position of the current object in the photo with the virtual image, and other photo positions of the other objects in the photo with the virtual image; According to the association relationship between the group photo position and the action feature, determine the current action feature associated with the current group photo position, and other action features associated with the other group photo positions; A current virtual image prototype corresponding to the current object is generated according to the current image feature and the current action feature, and other virtual image prototypes corresponding to the other objects are generated according to the other image features and the other action features.

28. The device according to claim 25, wherein The virtual image prototype generation module is further specifically used for: Generate a first description text according to the current image feature and the target clothing feature, and input the first description text into the image generation model; Determine a current virtual image prototype corresponding to the current object according to a first predicted virtual image prototype output by the image generation large model; Generate a second description text according to the other image features and the target clothing features, and input the second description text into the image generation model; The second predicted virtual image prototype output by the image generation model is used to determine other virtual image prototypes corresponding to the other objects.

29. The device according to claim 27, wherein: The virtual image prototype generation module is further specifically used for: Generate a third description text according to the current image feature and the current action feature, and input the third description text into the image generation model; Determine a current virtual image prototype corresponding to the current object according to a third predicted virtual image prototype output by the image generation large model; Generate a fourth description text according to the other image features and the other action features, and input the fourth description text into the image generation model; The other virtual image prototypes corresponding to the other objects are determined according to the fourth predicted virtual image prototype output by the image generation model.

30. The device according to claim 27, wherein: The virtual image group photo generation module is further specifically used for: Determine the number of subjects to be photographed, and the prototype size ratio between the current avatar prototype and the other avatar prototypes; Determine the current region coordinates of the current avatar prototype in the group photo according to the size of the group photo, the number of subjects in the group photo, the size ratio of the prototype and the current position in the group photo; Determine the coordinates of other regions where the other avatar prototypes are located in the group photo according to the size of the group photo, the number of the group photo subjects, the size ratio of the prototypes and the positions of the other group photos; A virtual image photo of the current object and the other objects is generated according to the current region coordinates and the current virtual image prototype, and the other region coordinates and the other virtual image prototypes.

31. The device according to claim 17, further comprising a group photo downloading module, specifically used for: According to the photo download request implemented by the current object, a target virtual image photo is determined from at least one historical virtual image photo; wherein, Each of the historical virtual image group photos is generated based on the current virtual image prototype and different numbers of other virtual image prototypes; The group photo downloading operation is performed on the group photo of the target virtual image.

32. The apparatus of claim 18 or 23, wherein: The image features to be configured include at least one of gender features to be configured, age features to be configured, appearance features to be configured, and style features to be configured.

33. An electronic device comprising: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 16.

34. A non-transitory computer-readable storage medium storing computer instructions, wherein: The computer instructions are used to cause the computer to execute the method according to any one of claims 1-16.

35. A computer program product comprising a computer program, which, when executed by a processor, implements the method according to any one of claims 1 to 16.

Images