Digital image action control method and device, electronic equipment and storage medium

Abstract

The invention provides a digital image action control method and device, electronic equipment and a storage medium, and the method comprises the steps: obtaining a digital image, the action control configuration information of the digital image and the description information of a virtual interaction entity, and obtaining the action control configuration information of the digital image and the description information of the virtual interaction entity according to the action control configuration information of the digital image and the description information of the virtual interaction entity; determining a digital image action control instruction to generate a cue word, generating the cue word according to the digital image action control instruction to determine a digital image action control instruction, and controlling the digital image to execute an interaction action with the virtual interaction entity according to the digital image action control instruction. According to the action control configuration information of the digital image and the description information of the virtual interaction entity, the digital image action control instruction is determined to generate the cue word, and then the digital image is controlled to execute the action according to the digital image action control instruction generated based on the cue word generated by the digital image action control instruction. The stability of the generated digital image can be improved.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of artificial intelligence technology, and in particular, to a digital image motion control method, apparatus, electronic device, and storage medium. Background Art

[0002] A digital image is a virtual character constructed through computer graphics and artificial intelligence technologies, and is widely used in fields such as virtual reality, film and television production, game interaction, and remote collaboration. With the rapid development of technologies such as virtual reality and digital twin, the digital image motion generation technology has become the core link to achieve highly realistic interaction.

[0003] For related digital image motion generation methods, mainly real-person videos or text descriptions are used as the input of the deep learning model, and the deep learning model directly outputs videos related to digital image motions.

[0004] However, it is difficult to ensure the consistency of digital image motions by the above methods. The coherence between frames will give people a false feeling, and at the same time, there will be situations affecting the visual perception such as local blurring and local color difference, which will further lead to low stability of the generated digital image motions. Summary of the Invention

[0005] In view of the above problems, a digital image motion control method, apparatus, electronic device, and storage medium are proposed to overcome the above problems or at least partially solve the above problems, including:

[0006] In a first aspect, the embodiments of the present application provide a digital image motion control method, and the method includes:

[0007] Obtain a digital image, the motion control configuration information of the digital image, and the description information of the virtual interaction entity;

[0008] Determine a digital image motion control instruction generation prompt word according to the motion control configuration information of the digital image and the description information of the virtual interaction entity;

[0009] Determine a digital image motion control instruction according to the digital image motion control instruction generation prompt word;

[0010] Control the digital image to execute an interaction action with the virtual interaction entity according to the digital image motion control instruction.

[0011] Optionally, the motion control configuration information of the digital image includes at least one of the motion description information of the digital image, the digital image motion control instruction template, and the spatial state parameters of the digital image.

[0012] Optionally, generating a prompt word based on the digital avatar motion control instruction to determine the digital avatar motion control instruction includes:

[0013] Inputting the generated prompt word of the digital avatar motion control instruction into a pre-generated large language model to obtain the digital avatar motion control instruction.

[0014] Optionally, after the step of controlling the digital avatar to perform an interaction action with the virtual interaction entity according to the digital avatar motion control instruction, the method further includes:

[0015] If the digital avatar successfully performs the action, store at least one of the action description information of the digital avatar, the digital avatar motion control instruction template, the spatial state parameters of the digital avatar, the description information of the virtual interaction entity, and the digital avatar motion control instruction.

[0016] Optionally, after the step of controlling the digital avatar to perform an interaction action with the virtual interaction entity according to the digital avatar motion control instruction, the method further includes:

[0017] If the digital avatar fails to perform the action, obtain the error information of the digital avatar motion control instruction;

[0018] Correct the digital avatar motion control instruction according to the error information to obtain the corrected digital avatar motion control instruction;

[0019] Control the digital avatar to perform the action according to the corrected digital avatar motion control instruction until the digital avatar successfully performs the action.

[0020] Optionally, correcting the digital avatar motion control instruction according to the error information to obtain the corrected digital avatar motion control instruction includes:

[0021] Obtain a digital avatar motion control instruction correction prompt word template;

[0022] Fill the error information, the digital avatar motion control instruction, the action description information of the digital avatar, the spatial state parameters of the digital avatar, the digital avatar motion control instruction template, and the description information of the virtual interaction entity into the digital avatar motion control instruction correction prompt word template to generate a digital avatar motion control instruction correction prompt word;

[0023] Determine the corrected digital avatar motion control instruction according to the digital avatar motion control instruction correction prompt word.

[0024] Optionally, determining the corrected digital image action control instruction according to the digital image action control instruction correction prompt word includes:

[0025] Inputting the digital image action control instruction correction prompt word into a pre-generated digital image action control instruction correction model to obtain the corrected digital image action control instruction.

[0026] In a second aspect, an embodiment of the present application provides a digital image action control device, and the device includes:

[0027] A data acquisition module, configured to acquire a digital image, action control configuration information of the digital image, and description information of a virtual interaction entity;

[0028] A prompt word construction module, configured to determine a digital image action control instruction generation prompt word according to the action control configuration information of the digital image and the description information of the virtual interaction entity;

[0029] A digital image action control instruction generation module, configured to determine a digital image action control instruction according to the digital image action control instruction generation prompt word;

[0030] A digital image action control module, configured to control the digital image to execute an interaction action with the virtual interaction entity according to the digital image action control instruction.

[0031] Optionally, the action control configuration information of the digital image includes at least one of action description information of the digital image, a digital image action control instruction template, and spatial state parameters of the digital image.

[0032] Optionally, the digital image action control instruction generation module includes:

[0033] A large language model processing sub-module, configured to input the digital image action control instruction generation prompt word into a pre-generated large language model to obtain a digital image action control instruction.

[0034] Optionally, the device further includes:

[0035] A digital image action-related information storage module, configured to store at least one of the action description information of the digital image, the digital image action control instruction template, the spatial state parameters of the digital image, the description information of the virtual interaction entity, and the digital image action control instruction if the digital image executes the action successfully.

[0036] Optionally, the device further includes:

[0037] An error message acquisition module for the instruction, which is used to acquire the error message of the digital image action control instruction if the execution of the action by the digital image fails;

[0038] An instruction correction module for correcting the digital image action control instruction according to the error message to obtain the corrected digital image action control instruction;

[0039] An instruction verification module for controlling the digital image to execute the action according to the corrected digital image action control instruction until the execution of the action by the digital image is successful.

[0040] Optionally, the instruction correction module includes:

[0041] A prompt word template acquisition sub-module for acquiring a digital image action control instruction correction prompt word template;

[0042] A data filling sub-module for filling the error message, the digital image action control instruction, the action description information of the digital image, the spatial state parameters of the digital image, the digital image action control instruction template, and the description information of the virtual interaction entity into the digital image action control instruction correction prompt word template to generate a digital image action control instruction correction prompt word;

[0043] A digital image action control instruction correction sub-module for determining the corrected digital image action control instruction according to the digital image action control instruction correction prompt word.

[0044] Optionally, the digital image action control instruction correction sub-module includes:

[0045] A digital image action control instruction correction model processing unit for inputting the digital image action control instruction correction prompt word into a pre-generated digital image action control instruction correction model to obtain the corrected digital image action control instruction.

[0046] In a third aspect, an embodiment of the present application further provides an electronic device, including: a processor; a memory for storing instructions executable by the processor, wherein the processor is configured to execute the instructions to implement the digital image action control method as described in any one of the above.

[0047] In a fourth aspect, an embodiment of the present application further provides a storage medium, when the instructions in the storage medium are executed by the processor of the electronic device, enabling the electronic device to execute the digital image action control method as described in any one of the above.

[0048] In an embodiment of the present application, digital images, action control configuration information of the digital images, and description information of virtual interaction entities are obtained. According to the action control configuration information of the digital images and the description information of the virtual interaction entities, a prompt word for generating a digital image action control instruction is determined. According to the prompt word for generating the digital image action control instruction, a digital image action control instruction is determined. The digital image is controlled to perform an interaction action with the virtual interaction entity according to the digital image action control instruction. In the present application, by determining the prompt word for generating the digital image action control instruction according to the action control configuration information of the digital images and the description information of the virtual interaction entities, and then generating the digital image action control instruction according to the prompt word for generating the digital image action control instruction to control the digital image to perform actions, this way of combining multiple pieces of information not only enables more reasonable planning and instruction basis for the control of the digital image actions, makes the whole process more interpretable, and is more conducive to ensuring the consistency and coherence of the digital image actions, but also controls the digital image to perform actions according to the digital image action control instruction, which can control the actions of the digital image more precisely and conveniently, and only needs to adjust the prompt word for generating the digital image action control instruction, thereby improving the stability of the generated digital image.

[0049] The above description is only an overview of the technical solution of the present application. In order to be able to understand the technical means of the present application more clearly, it can be implemented according to the content of the specification. And in order to make the above and other purposes, features and advantages of the present application more obvious and understandable, the following specific embodiments of the present application are specifically given. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] By reading the following detailed description of the preferred embodiments, various other advantages and benefits will become clear to those of ordinary skill in the art. The drawings are only for the purpose of showing the preferred embodiments and are not considered to be a limitation of the present application. And throughout the drawings, the same reference numerals are used to represent the same components. In the drawings:

[0051] Figure 1 is a step flowchart of a digital image action control method provided by an embodiment of the present application;

[0052] Figure 2 is a schematic flowchart of a digital image action control method provided by an embodiment of the present application;

[0053] Figure 3 is a block diagram of a digital image action control device provided by an embodiment of the present application;

[0054] Figure 4 is a structural diagram of an electronic device provided by an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0055] Exemplary embodiments of the present application will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present application are shown in the drawings, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided so that the present application can be more thoroughly understood and the scope of the present application can be fully conveyed to those skilled in the art.

[0056] For ease of understanding, the terms related to the embodiments of the present application will be briefly introduced first.

[0057] 1. Digital avatar

[0058] A digital avatar refers to a virtual human or character created through artificial intelligence technology. These digital avatars can simulate human appearance, behavior, speech, and emotions and are widely used in multiple scenarios.

[0059] 2. LLMs (Large Language Models)

[0060] Large Language Models are a type of deep neural network learning model with a large number of parameters. By processing a large amount of text data, large language models learn language patterns, grammar, and semantics, enabling them to understand and generate human language and playing an important role in the field of natural language processing (NLP). With the rapid development of current artificial intelligence technology, representative large language models such as GPT4, LLaMA, ERNIE Bot, and Doubao have shown increasingly powerful intelligent interaction capabilities.

[0061] 3. 3D engine

[0062] A 3D engine refers to a software framework for creating and rendering three-dimensional computer graphics, including components such as a renderer, a physical simulation system, and an animation system. A 3D engine can control the movement of a digital avatar because there is a three-dimensional coordinate system and transformation matrix to change the position, and at the same time, the animation system and script control can be associated with movement animations to achieve real movement in combination with physical simulation. Currently, the mainstream 3D engines include: Unreal Engine, Unity3D, and CryEngine, etc.

[0063] The following will, with reference to the accompanying drawings, through specific embodiments and their application scenarios, provide a detailed description of a digital avatar motion control method, apparatus, electronic device, and storage medium provided by the embodiments of the present application.

[0064] Figure 1 is a flowchart of the steps of a digital avatar motion control method provided by the embodiments of the present application. As Figure 1 shown, the method includes:

[0065] Step 101: Obtain the digital image, the action control configuration information of the digital image, and the description information of the virtual interaction entity.

[0066] Step 102: Determine the prompt for generating the digital image action control instruction based on the action control configuration information of the digital image and the description information of the virtual interaction entity.

[0067] Step 103: Determine the digital image action control instruction according to the prompt for generating the digital image action control instruction.

[0068] Step 104: Control the digital image to execute the interaction action with the virtual interaction entity according to the digital image action control instruction.

[0069] In some embodiments of the present application, as Figure 2 shown, the digital image can be designed in advance using a 3D engine according to the customer's requirements. The action control configuration information of the digital image includes at least one of the action description information of the digital image, the digital image action control instruction template, and the spatial state parameters of the digital image.

[0070] Among them, the action description information of the digital image refers to the text description of the digital image action provided by the user. For example, the action description information of the digital image can be "Move the digital image to the edge of the chair and sit down, and then pick up the guitar placed beside and play it."

[0071] The digital image action control instruction template refers to an example of the digital image action control instruction, which is used to enable the large language model to generate the digital image action control instruction according to the digital image action control instruction template. The preset digital image action control instruction template can be as follows.

[0072]

[0073] The spatial state parameters of the digital image include at least one of three-dimensional coordinates, orientation, rotation, scaling, animation data (such as: bone information, key frame information), collision data, rendering data (such as: texture, lighting change, etc.), and state data (such as: static, walking, acceleration, etc.). Among them, the collision data refers to the data on how the digital image and the environment interact. The preset spatial state parameters of the digital image can be as follows.

[0074] {"geometry":{"vertices":[...]}, / / Geometry: {"Vertices": [...]}

[0075] "material":{...}, / / Material: {...}

[0076] "animation": {"bones": [...], "animationKeyframes": [...],...}, / / Animation: {"bones": [...], "animationKeyframes": [...],...}

[0077] A virtual interactive entity refers to a virtual entity that has an interactive behavior with a digital avatar in the action description information of the digital avatar. For example, if the action description information of the digital avatar is "move the digital avatar to the edge of the chair and sit down, then pick up the guitar placed beside and play it", then the virtual interactive entities are "chair" and "guitar". The description information of the virtual interactive entity includes at least one of position, available operations, collision data, status data, rendering data, and animation data. The description information of the preset virtual interactive entity can be as follows.

[0078] Existing interactive virtual interactive entities: ["chair", "guitar"], and their specific information is as follows: {"chair": {"item_id": 1, 'desc': 'A chair that can be sat on', 'location': <...,...,...>, 'available_operations': ['sit','move', "lift"], "guitar": {"item_id": 2, 'desc': 'A guitar that can be played', 'location': <...,...,...>, 'available_operations': ['play','move', 'lift']}

[0079] The prompt words for generating digital avatar action control instructions can be used to make the large language model generate corresponding digital avatar action control instructions according to the prompt words for generating digital avatar action control instructions. As Figure 2 shown, after obtaining the action control configuration information of the digital avatar and the description information of the virtual interactive entity, construct the prompt words for generating digital avatar action control instructions. Specifically, the prompt words template for generating digital avatar action control instructions can be obtained in advance, and fill the action control configuration information of the digital avatar and the description information of the virtual interactive entity into the prompt words template for generating digital avatar action control instructions, then the prompt words for generating digital avatar action control instructions can be obtained. The prompt words for generating digital avatar action control instructions can be as follows.

[0080] You are a useful assistant for generating digital avatar action control instructions. Now you need to generate corresponding digital avatar action control instructions according to the action description information of the digital avatar.

[0081] 1. Action description information of the digital avatar

[0082] The action description information of the preset digital avatar is as follows:

[0083] Move the digital avatar to the edge of the chair and sit down, then pick up the guitar placed beside and play it.

[0084] 2. Spatial state parameters of the digital avatar

[0085] The preset spatial state parameters of the digital avatar are as follows:

[0086] {"geometry":{"vertices":[...]}, / / Geometry: {"vertices":[...]}

[0087] "material":{...}, / / Material: {...}

[0088] "animation":{"bones":[...]"animationKeyframes":[...]...}, / / Animation: {"bones":[...] animation keyframes":[...]...}

[0089] 3. Description information of virtual interactive entities:

[0090] The preset description information of virtual interactive entities is as follows:

[0091] Existing interactive virtual interactive entities: ["chair", "guitar"], and their specific information is as follows: {"chair": {"item_id": 1, 'desc': 'A chair that can be sat on', 'location': <...,...,...>, 'available_operations': ['sit','move', "lift"], 'guitar': {'item_id': 2, 'desc': 'A guitar that can be played', 'location': <...,...,...>, 'available_operations': ['play','move', 'lift']}

[0092] 4. Digital avatar action control instruction template

[0093] The following is the digital avatar action control instruction template:

[0094]

[0095]

[0096] After obtaining the prompt word, the digital image motion control instruction can be generated according to the digital image motion control instruction generation prompt word. Further, the digital image motion control instruction can be input into the 3D engine executor, and the digital image motion control instruction is run through the 3D engine executor. That is, the 3D engine executor can control the digital image to execute the interaction action with the virtual interaction entity according to the digital image motion control instruction. Among them, the 3D engine executor is the core component of the 3D engine and is used to run the digital image motion control instruction.

[0097] In this application, by determining the digital image motion control instruction generation prompt word according to the motion control configuration information of the digital image and the description information of the virtual interaction entity, and then generating the digital image motion control instruction according to the digital image motion control instruction generation prompt word to control the digital image to execute the action. This way of combining multiple pieces of information not only enables more reasonable planning and instruction basis for the control of the digital image motion, makes the whole process more interpretable, and is more conducive to ensuring the consistency and coherence of the digital image motion, but also can control the digital image to execute the action more precisely and conveniently according to the digital image motion control instruction, and only need to adjust the digital image motion control instruction generation prompt word, so as to improve the stability of the generated digital image. In addition,

[0098] Further, in some embodiments of this application, step 103 may further include the following steps:

[0099] Step 1031: Input the digital image motion control instruction generation prompt word into the pre-generated large language model to obtain the digital image motion control instruction.

[0100] In some embodiments of this application, the large language model is pre-trained according to the digital image motion control instruction generation prompt word and the digital image motion control instruction. The trained large language model has the ability to generate the digital image motion control instruction according to the digital image motion control instruction generation prompt word.

[0101] Specifically, as Figure 2 shown, after obtaining the digital image motion control instruction generation prompt word, the digital image motion control instruction generation prompt word can be used as the input data of the large language model and input into the large language model, and the large language model can automatically output the digital image motion control instruction.

[0102] In this application, by inputting the prompt words generated from the digital avatar action control instructions into the pre-trained large language model, the conversion from the prompt words generated from the digital avatar action control instructions to the digital avatar action control instructions is realized. This conversion is based on the pre-trained results of the large language model, and can quickly and accurately obtain the required digital avatar action control instructions, thereby improving the efficiency of generating digital avatar action control instructions.

[0103] Further, in some embodiments of this application, the method may further include the following steps:

[0104] Step 201, if the digital avatar successfully executes an action, store at least one of the action description information of the digital avatar, the digital avatar action control instruction template, the spatial state parameters of the digital avatar, the description information of the virtual interaction entity, and the digital avatar action control instruction.

[0105] In some embodiments of this application, as Figure 2 shown, if the 3D engine executor can successfully control the digital avatar to execute the interaction action with the virtual interaction entity according to the digital avatar action control instruction, that is, the digital avatar successfully executes the action, then save the information related to the digital avatar action. The information related to the digital avatar action includes at least one of the action description information of the digital avatar, the digital avatar action control instruction template, the spatial state parameters of the digital avatar, the description information of the virtual interaction entity, and the digital avatar action control instruction.

[0106] In this application, by saving at least one of the action description information of the digital avatar, the digital avatar action control instruction template, the spatial state parameters of the digital avatar, the description information of the virtual interaction entity, and the digital avatar action control instruction when the digital avatar successfully executes an action, not only can the accurate reproduction of the digital avatar action be realized, but also more optimized digital avatar action control strategies can be mined through further analysis of this information in the future, improving the fluency and naturalness of the digital avatar action.

[0107] Further, in some embodiments of this application, the method may further include the following steps:

[0108] Step 301, if the digital avatar fails to execute an action, obtain the error information of the digital avatar action control instruction.

[0109] Step 302, correct the digital avatar action control instruction according to the error information to obtain the corrected digital avatar action control instruction.

[0110] Step 303, control the digital avatar to execute the action according to the corrected digital avatar action control instruction until the digital avatar successfully executes the action.

[0111] In some embodiments of the present application, if the 3D engine actuator cannot control the digital avatar to perform an interaction action with the virtual interaction entity according to the digital avatar action control instruction, that is, the digital avatar fails to perform the action, an error message indicating the failure of the digital avatar action control instruction execution can be obtained. The error message indicating the failure of the digital avatar action control instruction execution can be as follows.

[0112]

[0113] After obtaining the error message indicating the failure of the digital avatar action control instruction execution, the digital avatar action control instruction can be optimized and adjusted according to the error message to obtain a corrected digital avatar action control instruction. The corrected digital avatar action control instruction is input into the 3D engine actuator again, and the corrected digital avatar action control instruction is run through the 3D engine actuator until the 3D engine actuator can successfully control the digital avatar to perform an interaction action with the virtual interaction entity according to the corrected digital avatar action control instruction, that is, the digital avatar successfully performs the action.

[0114] In the present application, when the digital avatar fails to perform the action, by timely obtaining the error message of the digital avatar action control instruction, the source that causes the failure of the digital avatar action execution can be accurately located in some parts of the digital avatar action control instruction, avoiding blindly conducting a large-scale investigation of the entire digital avatar action control instruction. Further, the efficiency of the successful execution of the digital avatar action is improved. In addition, by correcting the digital avatar action control instruction according to the error message, the accuracy of the digital avatar action control instruction can be improved, thereby increasing the accuracy rate of the successful execution of the digital avatar action.

[0115] Further, in some embodiments of the present application, step 302 may further include the following steps:

[0116] Step 401, obtain a template for the correction prompt word of the digital avatar action control instruction.

[0117] Step 402, fill the error message, the digital avatar action control instruction, the action description information of the digital avatar, the spatial state parameters of the digital avatar, the digital avatar action control instruction template, and the description information of the virtual interaction entity into the template for the correction prompt word of the digital avatar action control instruction to generate a correction prompt word for the digital avatar action control instruction.

[0118] Step 403, determine the corrected digital avatar action control instruction according to the correction prompt word of the digital avatar action control instruction.

[0119] In some embodiments of the present application, a digital avatar action control instruction correction prompt word template can be obtained in advance. After obtaining the digital avatar action control instruction correction prompt word template, at least one of the error information, the digital avatar action control instruction, the action description information of the digital avatar, the spatial state parameters of the digital avatar, the digital avatar action control instruction template, and the description information of the virtual interaction entity is filled into the digital avatar action control instruction correction prompt word template to obtain a digital avatar action control instruction correction prompt word. Among them, the digital avatar action control instruction correction prompt word can be used to enable the digital avatar action control instruction correction model to generate a corresponding corrected digital avatar action control instruction according to the digital avatar action control instruction correction prompt word. The digital avatar action control instruction correction prompt word can be as follows:

[0120] The following digital avatar action control instructions cannot run properly. Please repair them:

[0121] 1. Digital avatar action control instruction:

[0122]

[0123]

[0124] 2. Error information of the digital avatar action control instruction:

[0125] 3. Action description information of the digital avatar:

[0126] Move the digital avatar to the edge of the chair and sit down, then pick up the guitar placed beside and play it.

[0127] 4. Spatial state parameters of the digital avatar:

[0128] The preset spatial state parameters of the digital avatar are as follows: ...

[0130] 5. Description information of the virtual interaction entity:

[0131] Existing interactive items: ...

[0133] 6. Example of the corrected digital avatar action control instruction

[0134] The following is an example of the corrected digital avatar action control instruction: ...

[0136] After obtaining the digital avatar action control instruction correction prompt word, the corrected digital avatar action control instruction can be generated according to the digital avatar action control instruction correction prompt word.

[0137] In this application, by filling various information (error information, digital avatar action control instructions, action description information of the digital avatar, spatial state parameters of the digital avatar, digital avatar action control instruction templates, and description information of virtual interaction entities) into the digital avatar action control instruction correction prompt word template, a digital avatar action control instruction correction prompt word is generated, which can comprehensively consider various factors affecting the digital avatar action control instructions, so as to more accurately correct the problems existing in the current digital avatar action control instructions, improve the effectiveness and accuracy of the correction, thereby generating a more accurate digital avatar action control instruction correction prompt word, and improving the accuracy of the corrected digital avatar action control instructions.

[0138] Furthermore, in some embodiments of this application, step 403 may further include the following steps:

[0139] Step 4031, input the digital avatar action control instruction correction prompt word into a pre-generated digital avatar action control instruction correction model to obtain the corrected digital avatar action control instruction.

[0140] In some embodiments of this application, the digital avatar action control instruction correction model is pre-trained according to the digital avatar action control instruction correction prompt word and the corrected digital avatar action control instruction. The trained digital avatar action control instruction correction model has the ability to generate the corrected digital avatar action control instruction according to the digital avatar action control instruction correction prompt word.

[0141] Specifically, as Figure 2 shown, after obtaining the digital avatar action control instruction correction prompt word, the digital avatar action control instruction correction prompt word can be used as the input data of the digital avatar action control instruction correction model and input into the digital avatar action control instruction correction model, and the digital avatar action control instruction correction model can automatically output the corrected digital avatar action control instruction.

[0142] In this application, by inputting the digital avatar action control instruction correction prompt word into the pre-trained digital avatar action control instruction correction model, the model can automatically output the corrected digital avatar action control instruction, greatly reducing the time for manually checking and correcting the digital avatar action control instructions one by one, thereby improving the efficiency of obtaining the corrected digital avatar action control instructions.

[0143] Corresponding to the method provided in the above digital avatar action control method embodiment of this application, refer to Figure 3 , this application also provides a device block diagram of a digital avatar action control device. In this embodiment, the device includes:

[0144] The data acquisition module 301 is used to acquire digital avatars, action control configuration information of digital avatars, and description information of virtual interaction entities;

[0145] The prompt word construction module 302 is used to determine a prompt word for generating a digital avatar action control instruction according to the action control configuration information of the digital avatar and the description information of the virtual interaction entity;

[0146] The digital avatar action control instruction generation module 303 is used to determine a digital avatar action control instruction according to the prompt word for generating the digital avatar action control instruction;

[0147] The digital avatar action control module 304 is used to control the digital avatar to execute an interaction action with the virtual interaction entity according to the digital avatar action control instruction.

[0148] Optionally, the action control configuration information of the digital avatar includes at least one of action description information of the digital avatar, a digital avatar action control instruction template, and spatial state parameters of the digital avatar.

[0149] Optionally, the digital avatar action control instruction generation module 303 includes:

[0150] The large language model processing sub-module is used to input the prompt word for generating the digital avatar action control instruction into a pre-generated large language model to obtain a digital avatar action control instruction.

[0151] Optionally, the device further includes:

[0152] The digital avatar action-related information storage module is used to store at least one of the action description information of the digital avatar, the digital avatar action control instruction template, the spatial state parameters of the digital avatar, the description information of the virtual interaction entity, and the digital avatar action control instruction if the digital avatar executes the action successfully.

[0153] Optionally, the device further includes:

[0154] The error information acquisition module of the instruction is used to acquire the error information of the digital avatar action control instruction if the digital avatar executes the action failed;

[0155] The instruction correction module is used to correct the digital avatar action control instruction according to the error information to obtain a corrected digital avatar action control instruction;

[0156] The instruction verification module is used to control the digital avatar to execute the action according to the corrected digital avatar action control instruction until the digital avatar executes the action successfully.

[0157] Optionally, the instruction correction module includes:

[0158] A prompt template acquisition sub-module for acquiring a corrected prompt template for digital image action control instructions;

[0159] A data filling sub-module for filling error information, digital image action control instructions, action description information of the digital image, spatial state parameters of the digital image, a template for digital image action control instructions, and description information of the virtual interaction entity into the corrected prompt template for digital image action control instructions to generate a corrected prompt for digital image action control instructions;

[0160] A digital image action control instruction correction sub-module for determining the corrected digital image action control instructions according to the corrected prompt for digital image action control instructions.

[0161] Optionally, the digital image action control instruction correction sub-module includes:

[0162] A digital image action control instruction correction model processing unit for inputting the corrected prompt for digital image action control instructions into a pre-generated digital image action control instruction correction model to obtain the corrected digital image action control instructions.

[0163] In this application, by determining the prompt for generating digital image action control instructions based on the action control configuration information of the digital image and the description information of the virtual interaction entity, and then controlling the digital image to execute actions according to the digital image action control instructions generated from the prompt for generating digital image action control instructions, this multi-information combination method not only enables more reasonable planning and instruction basis for the control of digital image actions, makes the entire process more interpretable, and is more conducive to ensuring the consistency and coherence of digital image actions, but also can control the actions of the digital image more precisely and conveniently according to the digital image action control instructions, and only needs to adjust the prompt for generating digital image action control instructions, thereby improving the stability of the generated digital image.

[0164] Figure 4 It is a structural diagram of an electronic device M00 provided by an embodiment of this application. In the figure, the electronic device M00 includes a processor M01 and a memory M02. A program or instruction that can run on the processor M01 is stored on the memory M02. When the program or instruction is executed by the processor M01, it implements each step of the embodiment of the above digital image action control method and can achieve the same technical effect. To avoid repetition, it will not be elaborated here.

[0165] In an embodiment of the present application, the memory M02 can be used to store software programs and various data. The memory M02 mainly includes a first storage area for storing programs or instructions and a second storage area for storing data. Among them, the first storage area can store an operating system, application programs or instructions required for at least one function (such as a sound playback function, an image playback function, etc.). In addition, the memory M02 can include a volatile memory or a non-volatile memory, or the memory M02 can include both a volatile and a non-volatile memory. Among them, the non-volatile memory can be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory can be a random access memory (RAM), a static random access memory (SRAM), a dynamic random access memory (DRAM), a synchronous dynamic random access memory (SDRAM), a double data rate synchronous dynamic random access memory (DDR SDRAM), an enhanced synchronous dynamic random access memory (ESDRAM), a synchronous link dynamic random access memory (SLDRAM), and a direct rambus random access memory (DRRAM). The memory M02 in the embodiments of the present application includes, but is not limited to, these and any other suitable types of memories.

[0166] The processor M01 can include one or more processing units; optionally, the processor M01 integrates an application processor and a modem processor. Among them, the application processor mainly processes operations related to the operating system, user interface, and application programs, etc., and the modem processor mainly processes wireless communication signals, such as a baseband processor. It can be understood that the above modem processor may not be integrated into the processor M01 either.

[0167] The embodiments of the present application further provide a storage medium, on which a program or instruction is stored. When the program or instruction is executed by a processor, it implements each process of the above embodiment of the digital avatar action control method and can achieve the same technical effect. To avoid repetition, it will not be elaborated here.

[0168] Among them, the processor is the processor in the electronic device described in the above embodiments. The storage medium includes a computer-readable storage medium, such as a computer read-only memory ROM, a random access memory RAM, a magnetic disk, or an optical disc, etc.

[0169] Another embodiment of the present application provides a chip, which includes a processor and a communication interface. The communication interface is coupled to the processor. The processor is used to run programs or instructions to implement each process of the above embodiment of the digital avatar action control method, and can achieve the same technical effect. To avoid repetition, it will not be elaborated here.

[0170] Through the description of the above embodiments, those skilled in the art can clearly understand that the above embodiment methods can be implemented by means of software plus a necessary general hardware platform. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, in essence, or the part that contributes to the related technology, can be embodied in the form of a computer software product. The computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disc), and includes several instructions for causing a terminal (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present application.

[0171] The embodiments of the present application have been described above in conjunction with the accompanying drawings. However, the present application is not limited to the above specific implementation manners. The above specific implementation manners are merely illustrative and not restrictive. Under the inspiration of the present application, those of ordinary skill in the art can also make many forms without departing from the purpose of the present application and the scope protected by the claims, and all of them fall within the protection scope of the present application.

Claims

1. A digital image motion control method, characterized in that The method includes: Obtaining a digital avatar, action control configuration information of the digital avatar, and description information of a virtual interaction entity; Determining a prompt for generating a digital avatar action control instruction according to the action control configuration information of the digital avatar and the description information of the virtual interaction entity; Determining a digital avatar action control instruction according to the prompt for generating a digital avatar action control instruction; Controlling the digital avatar to perform an interaction action with the virtual interaction entity according to the digital avatar action control instruction.

2. The method according to claim 1, wherein The action control configuration information of the digital avatar includes at least one of action description information of the digital avatar, a digital avatar action control instruction template, and spatial state parameters of the digital avatar.

3. The method according to claim 1, characterized in that, The determining a digital avatar action control instruction according to the prompt for generating a digital avatar action control instruction includes: Inputting the prompt for generating a digital avatar action control instruction into a pre-generated large language model to obtain a digital avatar action control instruction.

4. The method according to claim 2, wherein After the step of controlling the digital avatar to perform an interaction action with the virtual interaction entity according to the digital avatar action control instruction, the method further includes: If the action of the digital avatar is successfully executed, storing at least one of the action description information of the digital avatar, the digital avatar action control instruction template, the spatial state parameters of the digital avatar, the description information of the virtual interaction entity, and the digital avatar action control instruction.

5. The method according to claim 2, characterized in that After the step of controlling the digital avatar to perform an interaction action with the virtual interaction entity according to the digital avatar action control instruction, the method further includes: If the action of the digital avatar fails, obtaining error information of the digital avatar action control instruction; Correcting the digital avatar action control instruction according to the error information to obtain a corrected digital avatar action control instruction; Controlling the digital avatar to perform an action according to the corrected digital avatar action control instruction until the action of the digital avatar is successfully executed.

6. The method according to claim 5, wherein The correcting the digital avatar action control instruction according to the error information to obtain a corrected digital avatar action control instruction includes: Obtaining a digital avatar action control instruction correction prompt word template; Filling the error information, the digital avatar action control instruction, the action description information of the digital avatar, the spatial state parameters of the digital avatar, the digital avatar action control instruction template, and the description information of the virtual interaction entity into the digital avatar action control instruction correction prompt word template to generate a digital avatar action control instruction correction prompt word; Determining a corrected digital avatar action control instruction according to the digital avatar action control instruction correction prompt word.

7. The method according to claim 6, wherein The determining a corrected digital avatar action control instruction according to the digital avatar action control instruction correction prompt word includes: Inputting the digital avatar action control instruction correction prompt word into a pre-generated digital avatar action control instruction correction model to obtain a corrected digital avatar action control instruction.

8. A digital image motion control device, characterized in that, The device includes: A data acquisition module, configured to acquire a digital avatar, the action control configuration information of the digital avatar, and the description information of a virtual interaction entity; A prompt word construction module, configured to determine a prompt word for generating a digital avatar action control instruction according to the action control configuration information of the digital avatar and the description information of the virtual interaction entity; A digital avatar action control instruction generation module, configured to determine a digital avatar action control instruction according to the prompt word for generating the digital avatar action control instruction; A digital avatar action control module, configured to control the digital avatar to execute an interaction action with the virtual interaction entity according to the digital avatar action control instruction.

9. An electronic device, characterized in that, Comprising: A processor; A memory for storing instructions executable by the processor; Wherein, the processor is configured to execute the instructions to implement the digital avatar action control method according to any one of claims 1 to 7.

10. A storage medium, characterized in that, When the instructions in the storage medium are executed by a processor of an electronic device, the electronic device is enabled to execute the digital avatar action control method according to any one of claims 1 to 7.

Images