Interactive content management method and device based on large model and electronic equipment thereof

Abstract

The invention provides an interactive content management method and device based on a large model and electronic equipment, and relates to the technical field of game interaction.The method comprises the steps that game picture frames of a plurality of users are obtained; wherein the game picture frame comprises user interaction content; performing feature extraction on all the game picture frames through the large model to obtain content feature vectors; wherein the content feature vector comprises an image feature vector and a text feature vector; performing feature fusion on the image feature vector and the text feature vector to obtain a fused feature vector; comparing the fusion feature vectors of different users one by one, and judging whether a target user group exists or not; wherein the target users comprise at least two users in the same plot node; and when the target user group exists, synchronizing the user interaction contents of all target users at the plot node, and displaying the user interaction contents through a game picture frame corresponding to the plot node. According to the invention, the accuracy and efficiency of content management are improved, and the cost and difficulty of manual auditing are reduced.

Description

Technical Field

[0001] The present invention relates to the field of game interaction technology, and in particular to a large model-based interactive content management method, device and electronic equipment. Background Art

[0002] In the current field of games and interactive media, user-generated content (UGC) such as bullet comments and reviews has become an important way for users to communicate and interact. These UGCs not only enrich the user's viewing experience, but also enhance the sense of connection and participation between users. However, in actual applications, differences in progress or inconsistent viewing time between users will lead to content synchronization problems, which seriously affects the user experience and content sharing effect.

[0003] Most existing content synchronization methods rely on timestamps or simple progress bar matching, which works well when users watch at similar speeds, but often fail to accurately synchronize content when there are large progress differences between users. For example, some users may choose to fast forward or rewind, while others may be playing normally, which will result in different plot contents seen by different users at the same time, making it impossible to effectively share each other's UGC.

[0004] Therefore, a method, device and electronic device for interactive content management based on a large model are proposed. Summary of the invention

[0005] This specification provides a method, device and electronic device for interactive content management based on a large model, which improves the accuracy and efficiency of content management and reduces the cost and difficulty of manual review.

[0006] This specification provides an interactive content management method based on a large model, including:

[0007] Acquire game screen frames of several users; wherein the game screen frames include user interaction content;

[0008] Extracting features from all the game screen frames through a large model to obtain a content feature vector; wherein the content feature vector includes an image feature vector and a text feature vector;

[0009] Performing feature fusion on the image feature vector and the text feature vector to obtain a fused feature vector;

[0010] Compare the fused feature vectors of different users one by one to determine whether there is a target user group; wherein the target users include at least two users at the same plot node;

[0011] When the target user group exists, the user interaction contents of all the target users are synchronized at the plot node and displayed through the game screen frame corresponding to the plot node.

[0012] Optionally, after obtaining the game screen frames of several users, the method further includes:

[0013] Analyze the user interaction content in the game screen frame through a large model to determine whether the user interaction content is compliant;

[0014] When the user interaction content is non-compliant, the non-compliant user interaction content is shielded in the game screen frame, and / or a warning prompt is sent to the user corresponding to the non-compliant user interaction content.

[0015] Optionally, after obtaining the game screen frames of several users, the method further includes:

[0016] The emotion recognition of the user interaction content in the game screen frame is performed through the large model to determine the emotion type of the game screen frame, and the user interaction content that matches the emotion type of the game screen frame is preferentially displayed at the plot node.

[0017] Optionally, after obtaining the game screen frames of several users, the method further includes:

[0018] Performing plot analysis on the game screen frame by using a large model to determine the emotion type of the game screen frame;

[0019] Based on the emotion type of the game screen frame, anthropomorphic interactive content that conforms to the emotion type of the game screen frame is automatically generated and displayed through the game screen frame.

[0020] Optionally, the determining whether the user interaction content is compliant includes:

[0021] Determining whether the user interactive content contains sensitive words;

[0022] and / or,

[0023] Determining whether the user interaction content contains negative emotions;

[0024] and / or,

[0025] Determine whether the user interactive content contains spoiler content.

[0026] Optionally, the extracting features of all the game screen frames by using a large model to obtain a content feature vector includes:

[0027] Extract features from all the game screen frames using a large model to obtain image features and text features;

[0028] The image features / the text features are compressed respectively to obtain an image feature vector and a text feature vector.

[0029] Optionally, also include:

[0030] The user interaction content is bound to the plot node and stored in a data block.

[0031] This specification provides an interactive content management device based on a large model, which is characterized by comprising:

[0032] An acquisition module, used to acquire game screen frames of several users; wherein the game screen frames include user interaction content;

[0033] An extraction module, used to extract features from all the game screen frames through a large model to obtain a content feature vector; wherein the content feature vector includes an image feature vector and a text feature vector;

[0034] A fusion module, used for fusing the image feature vector with the text feature vector to obtain a fused feature vector;

[0035] A judgment module, used to compare the fused feature vectors of different users one by one to determine whether there is a target user group; wherein the target users include at least two users at the same plot node;

[0036] A synchronization module is used to synchronize the user interaction content of all the target users at the plot node when the target user group exists, and display it through the game screen frame corresponding to the plot node.

[0037] Optionally, after the acquisition module, the following is included:

[0038] Analyze the user interaction content in the game screen frame through a large model to determine whether the user interaction content is compliant;

[0039] When the user interaction content is non-compliant, the non-compliant user interaction content is shielded in the game screen frame, and / or a warning prompt is sent to the user corresponding to the non-compliant user interaction content.

[0040] Optionally, after the acquisition module, the following further includes:

[0041] The emotion recognition of the user interaction content in the game screen frame is performed through the large model to determine the emotion type of the game screen frame, and the user interaction content that matches the emotion type of the game screen frame is preferentially displayed at the plot node.

[0042] Optionally, after the acquisition module, the following further includes:

[0043] Performing plot analysis on the game screen frame by using a large model to determine the emotion type of the game screen frame;

[0044] Based on the emotion type of the game screen frame, anthropomorphic interactive content that conforms to the emotion type of the game screen frame is automatically generated and displayed through the game screen frame.

[0045] Optionally, the judging module includes:

[0046] Determining whether the user interactive content contains sensitive words;

[0047] and / or,

[0048] Determining whether the user interaction content contains negative emotions;

[0049] and / or,

[0050] Determine whether the user interactive content contains spoiler content.

[0051] Optionally, the extraction module includes:

[0052] Extract features from all the game screen frames using a large model to obtain image features and text features;

[0053] The image features / the text features are compressed respectively to obtain an image feature vector and a text feature vector.

[0054] Optionally, also include:

[0055] The user interaction content is bound to the plot node and stored in a data block.

[0056] This specification also provides an electronic device, wherein the electronic device includes:

[0057] processor; and,

[0058] A memory storing computer executable instructions, which when executed cause the processor to perform any of the above methods.

[0059] The present specification also provides a computer-readable storage medium, wherein the computer-readable storage medium stores one or more programs, and when the one or more programs are executed by a processor, any of the above methods is implemented.

[0060] In the present invention, the game screen frame is deeply analyzed through a large model, which not only realizes the compliance judgment, emotion recognition and priority display of the user interactive content, but also can automatically generate anthropomorphic interactive content that meets the plot and emotional requirements. More importantly, this technical framework adopts a frame matching method, which enables it to seamlessly adapt to any game. Regardless of the type, style or technical architecture of the game, the frame matching technology can accurately capture every detail in the game screen, providing a solid foundation for subsequent compliance judgment, emotion recognition and interactive content generation. This wide compatibility means that whether it is a retro pixel-style game or a modern high-definition 3D masterpiece, the present invention can easily cope with it, ensuring that users can enjoy the same excellent interactive experience in different games. The use of frame matching technology not only ensures the accuracy and real-time nature of the analysis, but also greatly broadens the scope of application of the technology, so that game developers do not need to carry out customized development for each game, and can achieve efficient content management and personalized recommendations. In addition, the user interactive content can be bound and stored with the plot nodes, providing strong support for subsequent data analysis and personalized recommendations. Together, these features improve the user experience and interactivity of games and interactive media platforms, allowing users to more deeply integrate into the game world and enjoy a richer, more interesting and personalized gaming experience. At the same time, through the use of large models, this method also improves the accuracy and efficiency of content management and reduces the cost and difficulty of manual review. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0062] Figure 1 A schematic diagram of the principle of an interactive content management method based on a large model provided in an embodiment of this specification;

[0063] Figure 2 A schematic diagram of the structure of an interactive content management device based on a large model provided in an embodiment of this specification;

[0064] Figure 3 A schematic diagram of the structure of an electronic device provided in an embodiment of this specification;

[0065] Figure 4 A schematic diagram of a computer-readable medium provided for an embodiment of this specification. DETAILED DESCRIPTION

[0066] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art can think of other obvious variations. The basic principles of the present invention defined in the following description can be applied to other embodiments, variations, improvements, equivalents, and other technical solutions that do not deviate from the spirit and scope of the present invention.

[0067] The following is combined with Figure 1-4 The exemplary embodiments of the present invention are described more fully. However, the exemplary embodiments can be implemented in various forms, and it should not be understood that the present invention is limited to the embodiments set forth herein. On the contrary, providing these exemplary embodiments can make the present invention more comprehensive and complete, and it is more convenient to fully convey the inventive concept to those skilled in the art. The same reference numerals in the figures represent the same or similar elements, components or parts, and thus their repeated description will be omitted.

[0068] Under the premise of being consistent with the technical concept of the present invention, the features, structures, characteristics or other details described in a specific embodiment do not exclude that they can be combined in one or more other embodiments in a suitable manner.

[0069] In the description of specific embodiments, the features, structures, characteristics or other details described in the present invention are intended to enable those skilled in the art to fully understand the embodiments. However, it does not exclude that those skilled in the art can practice the technical solutions of the present invention without one or more of the specific features, structures, characteristics or other details.

[0070] The flowcharts shown in the accompanying drawings are only exemplary and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps can be decomposed, and some operations / steps can be combined or partially combined, so the actual execution order may change according to actual conditions.

[0071] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities may be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0072] The term "and / or" or "and / or" includes all combinations of any one or more of the associated listed items.

[0073] Figure 1 A schematic diagram of a large model-based interactive content management method provided in an embodiment of this specification, the method may include:

[0074] S110: Acquire game screen frames of several users; wherein the game screen frames include user interaction content;

[0075] In the specific implementation of this specification, user interactive content includes bullet screens, comments, etc. Large models include CLIP, GPT, etc. The GPT (Generative Pre-trained Transformer) model is a natural language generation model based on an autoregressive architecture. The autoregressive model predicts the output of the next moment based on the output of the previous moment, which is very suitable for sequential generation tasks, such as text generation. The core of the GPT model includes a multi-layer self-attention mechanism (Self-Attention) and a feed-forward neural network (Feed-forward Neural Networks), which enable the model to capture the contextual relationship between each word in a sentence. In the generation task, the GPT model starts with an initial input, gradually predicts the probability distribution of each word, and selects or samples the next word based on the probability. Since GPT is autoregressive, the generation result depends on the previously generated words, so the context relevance is strong. GPT uses unsupervised language model training during the training process, and gradually learns the grammatical, semantic, and syntactic features of the language by maximizing the probability of each word in the context.

[0076] On the other hand, the CLIP (Contrastive Language–Image Pre-training) model is a model for achieving multimodal alignment of images and texts. The core of CLIP is a dual encoder architecture, which includes an image encoder and a text encoder. The image encoder usually uses ResNet or Vision Transformer (ViT) to extract image features, while the text encoder uses a Transformer architecture similar to GPT to process text input. During the training process, CLIP uses a contrastive learning method to take image and corresponding text pairs as positive samples, and image and unrelated text pairs as negative samples, so as to maximize the similarity of image-text pairs and minimize the similarity between images and unpaired texts. Through contrastive loss, CLIP embeds similar images and texts closer in the vector space, giving the model stronger semantic alignment capabilities in visual and language understanding.

[0077] In practical applications, CLIP's multimodal alignment capability has been widely used in tasks such as image classification, search, and frame matching. In particular, in frame matching applications, CLIP can map images and descriptive text into a common embedding space to achieve high-precision image-text matching. This capability enables users to achieve semantically accurate recognition and comparison based on image content, making it possible to synchronize content across time periods.

[0078] S120: extracting features from all the game screen frames through a large model to obtain a content feature vector; wherein the content feature vector includes an image feature vector and a text feature vector;

[0079] Optionally, the S120 includes:

[0080] Extract features from all the game screen frames using a large model to obtain image features and text features;

[0081] The image features / the text features are compressed respectively to obtain an image feature vector and a text feature vector.

[0082] In a specific embodiment of the present specification, a pre-trained image recognition large model (such as a convolutional neural network CNN or a more advanced Vision Transformer, ViT) is used to analyze the game screen frame by frame. These models can capture detailed information in the picture, such as color, shape, texture, and the position and relationship of objects, thereby generating high-dimensional image features. These features not only contain the basic visual information of the picture, but also imply the semantic content of the picture. For the convenience and efficiency of subsequent processing, these high-dimensional image features are compressed. This process usually involves dimensionality reduction techniques, such as principal component analysis (PCA), linear discriminant analysis (LDA), or autoencoders in deep learning. Through these methods, high-dimensional image features can be converted into low-dimensional but information-rich image feature vectors, which can retain the main features of the original image while reducing computational complexity and storage requirements.

[0083] For the text content in the game screen (such as bullet comments, dialogues, prompt information, game titles, etc.), large models in the field of natural language processing (NLP) (such as BERT, GPT series, etc.) are used to extract text features. These models are trained with a large amount of text data and can deeply understand the semantics and contextual relationships of the text, thereby generating text features containing rich semantic information. Similarly, for the efficiency and convenience of subsequent processing, these high-dimensional text features are compressed. This process involves word embedding technology (such as Word2Vec, GloVe, etc.) or sentence embedding technology (such as Sentence-BERT, USE, etc.), which can convert text into low-dimensional, continuous vector representations. These vectors maintain the similarities and relationships between texts in the semantic space.

[0084] S130: Fusing the image feature vector with the text feature vector to obtain a fused feature vector;

[0085] In the specific implementation of this specification, the image feature vector and the text feature vector are preprocessed to ensure their consistency in terms of dimension, data format and numerical range. This includes but is not limited to normalization, standardization or alignment. There are many methods for feature fusion, common ones are concatenation, weighted summation, bilinear pooling, etc. The concatenation method is simple and direct, and the two vectors are directly connected together; the weighted summation performs weighted summation on the two vectors according to the preset weights or the weights obtained by learning; the bilinear pooling fuses the features by calculating the outer product of the two vectors and applying an appropriate pooling strategy. According to the selected fusion method, the image feature vector and the text feature vector are fused. For example, if the concatenation method is selected, the two vectors are concatenated in dimension; if the weighted summation is selected, the two vectors are weighted and summed according to the weights; if the bilinear pooling is selected, the outer product of the two vectors is calculated and the pooling strategy is applied. After the feature fusion, post-processing operations are performed to further refine and optimize the fused feature vector. This may include dimensionality reduction, normalization, or applying specific transformations, etc.

[0086] Through feature fusion, a fused feature vector containing both image information and text information can be obtained. This vector can play an important role in subsequent tasks such as frame matching, content synchronization, and UGC management, because it integrates information from both modalities, thereby improving the accuracy and efficiency of the task.

[0087] S140: Compare the fused feature vectors of different users one by one to determine whether there is a target user group; wherein the target users include at least two users at the same plot node;

[0088] In the specific implementation of this specification, the similarity of the fused feature vectors between each user is calculated by cosine similarity, Euclidean distance or dot product, and users meeting the similarity threshold are screened out. If the target user group may involve multiple subgroups, clustering (such as K-Means or DBSCAN) can be performed based on the vectors. It is determined whether there is a cluster close to the target feature in the clustering result.

[0089] After identifying the target user group, in UGC management, similar content or activities can be recommended based on the user group information; in terms of content synchronization, it can ensure that users at the same plot node can receive consistent game updates or plot advancement information; in social interaction, it can promote communication and cooperation within the user group, etc.

[0090] Optionally, the S140 includes:

[0091] Determining whether the user interactive content contains sensitive words;

[0092] and / or,

[0093] Determining whether the user interaction content contains negative emotions;

[0094] and / or,

[0095] Determine whether the user interactive content contains spoiler content.

[0096] In the specific implementation of this specification, based on content analysis, the system determines whether the user interaction content is compliant according to preset rules and models. This includes identifying non-compliant content such as sensitive words, potential spoilers, and negative emotions. For spoiler content, the system will also combine the user's current game progress to determine whether the barrage has leaked subsequent plots. When the system determines that the user's interactive content is non-compliant, it will immediately block the non-compliant content in the game screen frame to prevent it from affecting the experience of other users. The system will also send a warning prompt to the user corresponding to the non-compliant content, informing them that there is a problem with the content they posted, and reminding them to pay attention to compliance.

[0097] Determining whether the user interactive content contains spoiler content specifically includes: using a multimodal large model (such as LLaVA, BLIP-2, etc.) to jointly embed the game screen and bullet comments. Text and image features are aligned through a shared space to generate a semantic relevance score. Determine the relevance of the bullet comment content to the main events, characters, or task status in the current game screen. If the bullet comment reveals important information that cannot be directly inferred from the current screen (such as subsequent plots, hidden mechanisms, etc.), it is marked as a spoiler. Combined with the contextual video frame, consider whether the bullet comment content matches the game plot progression in a short period of time. Fine-tune the model through a pre-trained data set (including spoiler samples and non-spoiler samples) to improve the recognition ability of specific game plots and idioms. If a spoiler is detected, the system marks or intercepts the bullet comment, and provides an explanation of the reason for the spoiler detection. Bullet comments that are not judged as spoilers are displayed normally.

[0098] S150: When the target user group exists, the user interaction content of all the target users is synchronized at the plot node, and displayed through the game screen frame corresponding to the plot node.

[0099] In the specific implementation of this specification, each user can see the interactive content of other users at the same plot node, thereby enhancing the sociality and interactivity of the game.

[0100] Optionally, after S110, the following steps may be performed:

[0101] In a specific implementation of this specification, the user interaction content in the game screen frame is analyzed by a large model to determine whether the user interaction content is compliant;

[0102] When the user interaction content is non-compliant, the non-compliant user interaction content is shielded in the game screen frame, and / or a warning prompt is sent to the user corresponding to the non-compliant user interaction content.

[0103] Optionally, after S110, the step further includes:

[0104] In a specific implementation of the present specification, emotion recognition is performed on the user interaction content in the game screen frame through a large model to determine the emotion type of the game screen frame, and the user interaction content that matches the emotion type of the game screen frame is preferentially displayed at the plot node.

[0105] Match the emotional type of user interaction content with the atmosphere of the game screen frame. Give priority to barrage content that matches the current atmosphere, such as showing encouraging and caring barrages during tense plots, and showing humorous and relaxing barrages in relaxed scenes.

[0106] Optionally, after S110, the step further includes:

[0107] Performing plot analysis on the game screen frame by using a large model to determine the emotion type of the game screen frame;

[0108] Based on the emotion type of the game screen frame, anthropomorphic interactive content that conforms to the emotion type of the game screen frame is automatically generated and displayed through the game screen frame.

[0109] In the specific implementation of this specification, the multimodal large model is used to perform plot analysis on the game screen frame, including identifying the main characters, actions, environment and other elements of the current scene, as well as the relationship between them and the development of the storyline. Combining the results of the plot analysis with clues such as color, expression, action, etc. in the image, the system determines the emotion type of the game screen frame. Emotion types include happiness, sadness, tension, expectation, surprise, etc.

[0110] Based on the determined emotion type, the system uses the generation capability of a large multimodal model to automatically generate anthropomorphic interactive content that matches the emotion type of the current game screen frame. These contents can imitate the speaking style of real users, such as commenting on the plot, interacting with characters, or expressing emotions. For example, emotional comments can be generated in touching plots to express sympathy or support for the characters; and content of expectations and speculation can be generated in tense moments to guide other users to more deeply integrate into the plot.

[0111] The system will screen and optimize the anthropomorphic interactive content and display it in the form of virtual bullet screens in the game screen frame. These virtual bullet screens can be presented together with the real bullet screens of other users to form a richer interactive experience.

[0112] Optionally, also include:

[0113] The user interaction content is bound to the plot node and stored in a database.

[0114] In a specific implementation of the present specification, the collected user interaction content is bound to the identified plot nodes, which can be achieved by establishing an association relationship in a data structure, for example, creating an entry for each plot node in a data block and storing the related user interaction content as a sub-item of the entry.

[0115] In the present invention, the game screen frame is deeply analyzed through a large model, which not only realizes the compliance judgment, emotion recognition and priority display of the user interactive content, but also can automatically generate anthropomorphic interactive content that meets the plot and emotional requirements. More importantly, this technical framework adopts a frame matching method, which enables it to seamlessly adapt to any game. Regardless of the type, style or technical architecture of the game, the frame matching technology can accurately capture every detail in the game screen, providing a solid foundation for subsequent compliance judgment, emotion recognition and interactive content generation. This wide compatibility means that whether it is a retro pixel-style game or a modern high-definition 3D masterpiece, the present invention can easily cope with it, ensuring that users can enjoy the same excellent interactive experience in different games. The use of frame matching technology not only ensures the accuracy and real-time nature of the analysis, but also greatly broadens the scope of application of the technology, so that game developers do not need to carry out customized development for each game, and can achieve efficient content management and personalized recommendations. In addition, the user interactive content can be bound and stored with the plot nodes, providing strong support for subsequent data analysis and personalized recommendations. Together, these features improve the user experience and interactivity of games and interactive media platforms, allowing users to more deeply integrate into the game world and enjoy a richer, more interesting and personalized gaming experience. At the same time, through the use of large models, this method also improves the accuracy and efficiency of content management and reduces the cost and difficulty of manual review.

[0116] Figure 2 A schematic diagram of a large-model-based interactive content management device provided in an embodiment of this specification may include:

[0117] The acquisition module 10 is used to acquire game screen frames of several users; wherein the game screen frames include user interaction content;

[0118] An extraction module 20 is used to extract features from all the game screen frames through a large model to obtain a content feature vector; wherein the content feature vector includes an image feature vector and a text feature vector;

[0119] A fusion module 30, used for performing feature fusion on the image feature vector and the text feature vector to obtain a fused feature vector;

[0120] A judgment module 40 is used to compare the fused feature vectors of different users one by one to determine whether there is a target user group; wherein the target users include at least two users at the same plot node;

[0121] The synchronization module 50 is used to synchronize the user interaction content of all the target users at the plot node when the target user group exists, and display it through the game screen frame corresponding to the plot node.

[0122] Optionally, after the acquisition module 10, the following steps are included:

[0123] Analyze the user interaction content in the game screen frame through a large model to determine whether the user interaction content is compliant;

[0124] When the user interaction content is non-compliant, the non-compliant user interaction content is shielded in the game screen frame, and / or a warning prompt is sent to the user corresponding to the non-compliant user interaction content.

[0125] Optionally, after the acquisition module 10, the following further comprises:

[0126] The emotion recognition of the user interaction content in the game screen frame is performed through the large model to determine the emotion type of the game screen frame, and the user interaction content that matches the emotion type of the game screen frame is preferentially displayed at the plot node.

[0127] Optionally, after the acquisition module 10, the following further comprises:

[0128] Performing plot analysis on the game screen frame by using a large model to determine the emotion type of the game screen frame;

[0129] Based on the emotion type of the game screen frame, anthropomorphic interactive content that conforms to the emotion type of the game screen frame is automatically generated and displayed through the game screen frame.

[0130] Optionally, the judging module 40 includes:

[0131] Determining whether the user interactive content contains sensitive words;

[0132] and / or,

[0133] Determining whether the user interaction content contains negative emotions;

[0134] and / or,

[0135] Determine whether the user interactive content contains spoiler content.

[0136] Optionally, the extraction module 20 includes:

[0137] Extract features from all the game screen frames using a large model to obtain image features and text features;

[0138] The image features / the text features are compressed respectively to obtain an image feature vector and a text feature vector.

[0139] Optionally, also include:

[0140] The user interaction content is bound to the plot node and stored in a data block.

[0141] The functions of the device in the embodiment of the present invention have been described in the above method embodiment, so for details not provided in the description of this embodiment, please refer to the relevant description in the above embodiment, and no further description will be given here.

[0142] Based on the same inventive concept, an embodiment of this specification also provides an electronic device.

[0143] The following describes an electronic device embodiment of the present invention, which can be regarded as a specific physical implementation of the method and device embodiments of the present invention. The details described in the electronic device embodiment of the present invention should be regarded as a supplement to the above method or device embodiments; details not disclosed in the electronic device embodiment of the present invention can be implemented with reference to the above method or device embodiments.

[0144] Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this specification. Figure 3 The electronic device 300 according to the embodiment of the present invention is described. Figure 3 The electronic device 300 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0145] like Figure 3 As shown, the electronic device 300 is in the form of a general computing device. The components of the electronic device 300 may include, but are not limited to: at least one processing unit 310, at least one storage unit 320, a bus 330 connecting different system components (including the storage unit 320 and the processing unit 310), a display unit 340, etc.

[0146] The storage unit stores program codes, which can be executed by the processing unit 310, so that the processing unit 310 performs the steps according to various exemplary embodiments of the present invention described in the above processing method section of this specification. For example, the processing unit 310 can perform the following steps: Figure 1 Steps shown.

[0147] The storage unit 320 may include a readable medium in the form of a volatile storage unit, such as a random access memory unit (RAM) 3201 and / or a cache memory unit 3202 , and may further include a read-only memory unit (ROM) 3203 .

[0148] The storage unit 320 may also include a program / utility 3204 having a set (at least one) of program modules 3205, such program modules 3205 including but not limited to: an operating system, one or more application programs, other program modules and program data, each of which or some combination may include the implementation of a network environment.

[0149] Bus 330 may represent one or more of several types of bus structures, including a memory unit bus or memory unit controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus architectures.

[0150] The electronic device 300 may also communicate with one or more external devices 400 (e.g., keyboards, pointing devices, Bluetooth devices, etc.), one or more devices that enable viewers to interact with the electronic device 300, and / or any device that enables the electronic device 300 to communicate with one or more other computing devices (e.g., routers, modems, etc.). Such communication may be performed through an input / output (I / O) interface 350. Furthermore, the electronic device 300 may also communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) through a network adapter 360. The network adapter 360 may communicate with other modules of the electronic device 300 through the bus 330. It should be understood that although Figure 3 Not shown, other hardware and / or software modules may be used in conjunction with the electronic device 300, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0151] Through the description of the above implementation methods, it is easy for those skilled in the art to understand that the exemplary embodiments described in the present invention can be implemented by software, or by software combined with necessary hardware. Therefore, the technical solution according to the implementation method of the present invention can be embodied in the form of a software product, which can be stored in a computer-readable storage medium (which can be a CD-ROM, USB flash drive, mobile hard disk, etc.) or on a network, including a number of instructions to enable a computing device (which can be a personal computer, server, or network device, etc.) to execute the above method according to the present invention. When the computer program is executed by a data processing device, the computer-readable medium can implement the above method of the present invention, that is: Figure 1 The method shown.

[0152] Figure 4 A schematic diagram of a computer-readable medium provided for an embodiment of this specification.

[0153] accomplish Figure 1 The computer program of the method shown can be stored on one or more computer readable media. The computer readable medium can be a readable signal medium or a readable storage medium. The readable storage medium can be, for example, but not limited to, a system, device or device of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination of the above. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable 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 above.

[0154] The computer readable storage medium may include a data signal propagated in a baseband or as part of a carrier wave, wherein a readable program code is carried. This propagated data signal may take a variety of forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above. The readable storage medium may also be any readable medium other than a readable storage medium, which may send, propagate, or transmit a program for use by an instruction execution system, an apparatus, or a device or used in combination with it. The program code contained on the readable storage medium may be transmitted with any appropriate medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination of the above.

[0155] Program code for performing the operations of the present invention may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, etc., and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the viewer computing device, partially on the viewer device, as a stand-alone software package, partially on the viewer computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving a remote computing device, the remote computing device may be connected to the viewer computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0156] In summary, the present invention can be implemented in hardware, or in a software module running on one or more processors, or in a combination thereof. It should be understood by those skilled in the art that general data processing devices such as microprocessors or digital signal processors (DSPs) can be used in practice to implement some or all of the functions of some or all of the components in the embodiments of the present invention. The present invention can also be implemented as a device or apparatus program (e.g., a computer program and a computer program product) for executing part or all of the methods described herein. Such a program implementing the present invention can be stored on a computer-readable medium, or can have the form of one or more signals. Such a signal can be downloaded from an Internet website, or provided on a carrier signal, or provided in any other form.

[0157] The specific embodiments described above further describe the purpose, technical solutions and beneficial effects of the present invention in detail. It should be understood that the present invention is not inherently related to any specific computer, virtual device or electronic device, and various general devices can also implement the present invention. The above description is only a specific embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

[0158] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0159] The above is only the embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims

1. A method for interactive content management based on a large model, characterized in that: include: Acquire game screen frames of several users; wherein the game screen frames include user interaction content; Extracting features from all the game screen frames through a large model to obtain a content feature vector; wherein the content feature vector includes an image feature vector and a text feature vector; Performing feature fusion on the image feature vector and the text feature vector to obtain a fused feature vector; Compare the fused feature vectors of different users one by one to determine whether there is a target user group; wherein the target users include at least two users at the same plot node; When the target user group exists, the user interaction contents of all the target users are synchronized at the plot node and displayed through the game screen frame corresponding to the plot node.

2. The interactive content management method based on a large model as claimed in claim 1, characterized in that: After obtaining the game screen frames of several users, the method includes: Analyze the user interaction content in the game screen frame through a large model to determine whether the user interaction content is compliant; When the user interaction content is non-compliant, the non-compliant user interaction content is shielded in the game screen frame, and / or a warning prompt is sent to the user corresponding to the non-compliant user interaction content.

3. The interactive content management method based on a large model as claimed in claim 2, characterized in that: After obtaining the game screen frames of several users, the method further includes: The emotion recognition of the user interaction content in the game screen frame is performed through the large model to determine the emotion type of the game screen frame, and the user interaction content that matches the emotion type of the game screen frame is preferentially displayed at the plot node.

4. The interactive content management method based on a large model as claimed in claim 3, characterized in that: After obtaining the game screen frames of several users, the method further includes: Performing plot analysis on the game screen frame by using a large model to determine the emotion type of the game screen frame; Based on the emotion type of the game screen frame, anthropomorphic interactive content that conforms to the emotion type of the game screen frame is automatically generated and displayed through the game screen frame.

5. The interactive content management method based on a large model as claimed in claim 4, characterized in that: The determining whether the user interaction content is compliant includes: Determining whether the user interactive content contains sensitive words; and / or, Determining whether the user interaction content contains negative emotions; and / or, Determine whether the user interactive content contains spoiler content.

6. The interactive content management method based on a large model as claimed in claim 5, characterized in that: The feature extraction of all the game screen frames by the large model to obtain the content feature vector includes: Extract features from all the game screen frames using a large model to obtain image features and text features; The image features / the text features are compressed respectively to obtain an image feature vector and a text feature vector.

7. The interactive content management method based on a large model as claimed in claim 6, characterized in that: Also includes: The user interaction content is bound to the plot node and stored in a data block.

8. An interactive content management device based on a large model, characterized in that: include: An acquisition module, used to acquire game screen frames of several users; wherein the game screen frames include user interaction content; An extraction module, used to extract features from all the game screen frames through a large model to obtain a content feature vector; wherein the content feature vector includes an image feature vector and a text feature vector; A fusion module, used for fusing the image feature vector with the text feature vector to obtain a fused feature vector; A judgment module, used to compare the fused feature vectors of different users one by one to determine whether there is a target user group; wherein the target users include at least two users at the same plot node; A synchronization module is used to synchronize the user interaction content of all the target users at the plot node when the target user group exists, and display it through the game screen frame corresponding to the plot node.

9. An electronic device, wherein: The electronic device includes: processor; and, A memory storing computer executable instructions which, when executed, cause the processor to perform a method according to any one of claims 1-7.

10. A computer-readable storage medium, wherein: The computer-readable storage medium stores one or more programs, and when the one or more programs are executed by a processor, the method of any one of claims 1 to 7 is implemented.

Images