Multimedia resource processing method and device, electronic equipment and storage medium
Patent Information
- Application Number
- CN202510141266.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-02-08
- Publication Date
- 2026-08-11
Smart Images
Figure CN122547996A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of multimedia resource technology, and in particular to a multimedia resource processing method, apparatus, electronic device and storage medium. Background Technology
[0002] In the cold start scenario of multimedia resources, the existing technology requires operators to generate corresponding multimedia resources for multiple domain directions. This multimedia resource generation scheme relies on a lot of manual labor and cannot achieve the generation of large-scale multimedia resources. As a result, the generation efficiency of multimedia resources in the cold start scenario is low, and the amount of multimedia resources is sparse. Summary of the Invention
[0003] The technical problem to be solved by this application is to provide a multimedia resource processing method, apparatus, electronic device and storage medium, which can improve the generation efficiency of multimedia resources and expand the amount of multimedia resources in the case of multimedia resource cold start.
[0004] To address the aforementioned technical problems, this application provides a multimedia resource processing method, comprising:
[0005] Obtain initial multimedia resources;
[0006] Based on the association information between the resource content of the initial multimedia resource and multiple resource categories in the target publishing platform, the content format of the initial multimedia resource is expanded to obtain a target multimedia resource corresponding to the target resource category; the target resource category is the resource category among the multiple resource categories that is associated with the resource content of the initial multimedia resource, and the content format of the target multimedia resource matches the target resource category;
[0007] The target multimedia resource is published under the target resource category.
[0008] On the other hand, this application provides a multimedia resource processing apparatus, including:
[0009] The resource acquisition module is used to acquire initial multimedia resources;
[0010] The resource processing module is used to expand the content format of the initial multimedia resource based on the resource content of the initial multimedia resource and the association information of multiple resource categories in the target publishing platform, so as to obtain a target multimedia resource corresponding to the target resource category; the target resource category is the resource category associated with the resource content of the initial multimedia resource among the multiple resource categories, and the content format of the target multimedia resource matches the target resource category;
[0011] The resource publishing module is used to publish the target multimedia resources under the target resource category.
[0012] On the other hand, this application provides an electronic device including a processor and a memory, wherein the memory stores at least one instruction or at least one program, the at least one instruction or the at least one program being loaded and executed by the processor to implement the multimedia resource processing method described above.
[0013] On the other hand, this application provides a computer storage medium storing at least one instruction or at least one program, wherein the at least one instruction or the at least one program is loaded by a processor and executed as described above in the multimedia resource processing method.
[0014] On the other hand, this application provides a computer program product including computer instructions stored in a computer-readable storage medium; a processor of an electronic device reads the computer instructions from the computer-readable storage medium and executes the computer instructions to cause the electronic device to perform the above-described multimedia resource processing method.
[0015] Implementing the embodiments of this application has the following beneficial effects:
[0016] This application can determine target multimedia resources associated with the content of initial multimedia resources based on the association information between the resource content and multiple resource categories in the target publishing platform. Then, by expanding the content format of the initial multimedia resources, target multimedia resources corresponding to the target resource category can be obtained. In other words, this application can automatically associate and match initial multimedia resources with multiple resource categories. The corresponding initial multimedia resources can be used to expand the content format of multimedia resources in their associated target resource categories, thus eliminating the need for manual generation of multimedia resources for each resource category. The initial multimedia resources can be matched and used by multiple resource categories, thereby improving the utilization rate of the initial multimedia resources. Furthermore, automatically generating corresponding multimedia resources for multiple resource categories using the initial multimedia resources improves the efficiency of generating multimedia resources for multiple resource categories and expands the amount of multimedia resources available. Attached Figure Description
[0017] To more clearly illustrate the technical solutions and advantages in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0018] Figure 1 This is a schematic diagram of the implementation environment provided in the embodiments of this application;
[0019] Figure 2 This is a flowchart of a multimedia resource processing method provided in an embodiment of this application;
[0020] Figure 3 This is a flowchart of a method for generating multimedia resources based on associated information provided in an embodiment of this application;
[0021] Figure 4 This is a flowchart of a method for generating target multimedia resources based on an image adaptation model, provided in an embodiment of this application.
[0022] Figure 5 This is a schematic diagram of the joint structure of the image adaptation model and the text-based image model provided in the embodiments of this application;
[0023] Figure 6 This is a flowchart of a resource publishing method provided in an embodiment of this application;
[0024] Figure 7 This is a schematic diagram of the multimedia resource processing framework provided in the embodiments of this application;
[0025] Figure 8 This is a schematic diagram of a multimedia resource processing device provided in an embodiment of this application;
[0026] Figure 9 This is a schematic diagram of an electronic device structure provided in an embodiment of this application. Detailed Implementation
[0027] To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this application, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.
[0028] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or server that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.
[0029] In this application embodiment, the terms "module" or "unit" refer to a computer program or part of a computer program that has a predetermined function and works with other related parts to achieve a predetermined goal, and can be implemented wholly or partially using software, hardware (such as processing circuitry or memory), or a combination thereof. Similarly, a processor (or multiple processors or memory) can be used to implement one or more modules or units. Furthermore, each module or unit can be part of an overall module or unit that includes the functionality of that module or unit.
[0030] It is understood that in the specific embodiments of this application, data such as user information are involved. When the above embodiments of this application are applied to specific products or technologies, user permission or consent is required, and the collection, use and processing of related data must comply with the relevant laws, regulations and standards of the relevant countries and regions.
[0031] Please see Figure 1 The illustration shows an implementation environment provided in the embodiments of this application. The implementation environment may include at least one target terminal 110 and a resource server 120, and the target terminal 110 and the resource server 120 may communicate with each other via a network.
[0032] Specifically, the target terminal 110 can generate multimedia resources and publish them to the resource server 120; the target terminal 110 can also obtain multimedia resources from the resource server 120 and display them on the target terminal 110. The resource server 120 can store the multimedia resources generated by each target terminal and distribute them to the target terminals; furthermore, the resource server 120 can also generate multimedia resources. For example, in a cold start scenario, the resource server 120 can expand the content format based on the initial multimedia resources to automatically generate target multimedia resources and distribute them.
[0033] The target terminal 110 can communicate with the resource server 120 based on a browser / server (B / S) or client / server (C / S) model. The target terminal 110 may include physical devices such as smartphones, tablets, laptops, digital assistants, smart wearable devices, in-vehicle terminals, and servers, and may also include software running on the physical device, such as applications. The operating system running on the target terminal 110 in this embodiment may include, but is not limited to, Android, iOS, Linux, and Windows.
[0034] Resource server 120 and target terminal 110 can establish a communication connection via wired or wireless means. Resource server 120 may include a stand-alone server, a distributed server, or a server cluster consisting of multiple servers, wherein the server may be a cloud server.
[0035] To address the issues of low efficiency in generating multimedia resources and sparse multimedia resource quantity in cold start scenarios in existing technologies, this application provides a multimedia resource processing method. The execution entity of this method can be the aforementioned resource server. Please refer to [link to relevant documentation]. Figure 2 The method may include:
[0036] S210. Obtain initial multimedia resources.
[0037] In this embodiment, the initial multimedia resource can be a multimedia resource idea or a multimedia resource theme, or it can be a multimedia resource that can be directly published. The initial multimedia resource can serve as the basis for generating subsequent multimedia resources; that is, by expanding the content format based on the initial multimedia resource, the corresponding multimedia resource can be generated. Specifically, obtaining the initial multimedia resource includes:
[0038] Obtain historical multimedia resources published by the user account, resource search information of the user account, and popular multimedia resources;
[0039] Based on the historical multimedia resources, the resource search information, and the hot multimedia resources, resource analysis is performed to obtain the initial multimedia resources.
[0040] A user account can be the account used by a user to log in to a target publishing platform based on a target terminal. The target publishing platform can be the publishing platform corresponding to a resource server. Historical multimedia resources published by the user account can be obtained, along with the user account's presence on the target publishing platform and in popular multimedia resources. By analyzing the obtained information, initial multimedia resources can be obtained. In this embodiment, resource analysis based on the historical multimedia resources, the resource search information, and the popular multimedia resources can specifically include: extracting resource themes from historical multimedia resources, resource search information, and popular multimedia resources to obtain target resource themes; expanding the target resource themes with related themes to obtain related resource themes; and acquiring related resources from historical multimedia resources, resource search information, and popular multimedia resources to obtain related resources, etc. Accordingly, initial multimedia resources can be determined based on at least one of the target resource themes, related resource themes, and related resources.
[0041] In this embodiment, the initial multimedia resources can be determined by multiple dimensions of information, such as published historical multimedia resources, resource search information, and hot multimedia resources. This can expand the channels for obtaining the initial multimedia resources and improve the richness and quantity of the initial multimedia resources.
[0042] S220. Based on the association information between the resource content of the initial multimedia resource and multiple resource categories in the target publishing platform, the content format of the initial multimedia resource is expanded to obtain a target multimedia resource corresponding to the target resource category; the target resource category is the resource category among the multiple resource categories that is associated with the resource content of the initial multimedia resource, and the content format of the target multimedia resource matches the target resource category.
[0043] In this embodiment, multiple resource categories in the target publishing platform can refer to multiple publishing channels or multiple content tags in the target publishing platform. For example, multiple resource categories in the target publishing platform may include health, wallpapers, movies and TV series, comics, lifestyle, etc.
[0044] In this embodiment, an initial multimedia resource may be used by multiple resource categories, thereby enabling the generation of multimedia resources adapted to multiple resource categories for a single initial multimedia resource. Specifically, the resource content of the initial multimedia resource can be associated and matched with multiple resource categories in the target publishing platform to obtain association information between the resource content of the initial multimedia resource and multiple resource categories. The association information can include association and non-association; multiple resource categories can be associated with all of the initial multimedia resource; some resource categories can be associated with the initial multimedia resource, while the remaining resource categories are not associated with the initial multimedia resource. Correspondingly, multimedia resources can be generated for the resource categories associated with the initial multimedia resource, while no processing is performed on the resource categories not associated with the initial multimedia resource. For example, multiple resource categories in the target publishing platform include resource category 1, resource category 2, resource category 3, resource category 4, etc. The resource content of the initial multimedia resource is associated with resource category 1 and resource category 2, thereby enabling the generation of target multimedia resources matching resource category 1 and resource category 2 based on the initial multimedia resource.
[0045] Furthermore, the expansion of the content format of multimedia resources can include format transformation expansion between different content formats, as well as content expansion between the same formats. Format transformation expansion between different content formats can include converting content format from text to image, or from image to text. Content expansion between the same formats can include text rewriting, image-to-image generation, etc., where text rewriting can be performed on text-based content to obtain text-based content, and image-to-image generation can be generated from image-based content. Correspondingly, expanding the content format of the initial multimedia resources can involve performing at least one of the following operations: text-to-image, image-to-text, text rewriting, or image-to-image generation.
[0046] It should be noted that multimedia resources generated based on content format expansion can be published directly as target multimedia resources, or they can be combined with other multimedia resources and then published as target multimedia resources.
[0047] S230. Publish the target multimedia resource under the target resource category.
[0048] For each target resource category, the corresponding target multimedia resources can be published under the corresponding target resource category, which facilitates management and retrieval by category.
[0049] This application can determine target multimedia resources associated with the content of initial multimedia resources based on the association information between the resource content and multiple resource categories in the target publishing platform. Then, by expanding the content format of the initial multimedia resources, target multimedia resources corresponding to the target resource category can be obtained. In other words, this application can automatically associate and match initial multimedia resources with multiple resource categories. The corresponding initial multimedia resources can be used to expand the content format of multimedia resources in their associated target resource categories, thus eliminating the need for manual generation of multimedia resources for each resource category. The initial multimedia resources can be matched and used by multiple resource categories, thereby improving the utilization rate of the initial multimedia resources. Furthermore, automatically generating corresponding multimedia resources for multiple resource categories using the initial multimedia resources improves the efficiency of generating multimedia resources for multiple resource categories and expands the amount of multimedia resources available.
[0050] Specifically, in this embodiment, the association information between the initial multimedia resource and multiple resource categories can be determined based on the resource content of the initial multimedia resource and the resource characteristics of multiple resource categories; for details, please refer to... Figure 3 It illustrates a method for generating multimedia resources based on associated information, which may include:
[0051] S310. Obtain the resource features corresponding to each of the multiple resource categories; the resource features corresponding to each resource category include resource content features and content form features.
[0052] The resource content features of each resource category can characterize the theme information of the multimedia resources under that resource category. Specifically, the resource content features can be related to theme A, related to theme B, or have no theme limitation. The content form features of each resource category can characterize the content form of the multimedia resources under that resource category. Specifically, the content form features can be content form a, content form b, or have no content form limitation.
[0053] S320. Match the resource content features of the multiple resource categories with the resource content of the initial multimedia resource, and determine the target resource category from the multiple resource categories based on the content matching results; the resource content features of the target resource category are matched with the resource content of the initial multimedia resource.
[0054] First, the resource content features of multiple resource categories can be matched with the resource content of the initial multimedia resource to obtain content matching results. The content matching results can include content matching and content mismatch. Among them, the resource content features of the target resource category match the resource content of the initial multimedia resource, while the resource content features of the resource categories other than the target resource category do not match the resource content of the initial multimedia resource.
[0055] S330. Based on the content format characteristics of the target resource classification, the initial multimedia resource is expanded in terms of content format to obtain the target multimedia resource.
[0056] Different resource categories may have different content format characteristics. Accordingly, the content format of the initial multimedia resources can be expanded based on the content format characteristics of the target resource category to obtain target multimedia resources that match the content format characteristics of the target resource category.
[0057] The target publishing platform can include multiple resource categories such as health, wallpapers, movies and TV series, and games. For the health resource category, its resource content characteristics include health-related themes, and its content format is not limited, and can be text, images, etc. For the wallpaper resource category, its resource content characteristics are not limited, and its content format is images. For the movie and TV series resource category, its resource content characteristics are movie and TV series-related themes, and its content format is not limited, and can be text, images, etc. For the game resource category, its resource content characteristics include game-related themes, and its content format is not limited, and can be text, images, etc.
[0058] Specifically, for example, if the content of the initial multimedia resource is related to food 1, then the corresponding health resource category can involve topics related to food 1, such as an introduction to food 1, healthy cooking methods for food 1, etc.; the wallpaper resource category can involve images containing food 1, etc., thus determining that the health resource category and the wallpaper resource category can be target resource categories that match the content of the initial resource; furthermore, the content form of the initial multimedia resource can be expanded based on the content form characteristics of the health resource category and the wallpaper resource category to obtain target multimedia resources corresponding to the health resource category and target multimedia resources corresponding to the wallpaper resource category.
[0059] In this embodiment, each resource category has corresponding resource content features and content form features. Therefore, a target resource category that matches the initial multimedia resource can be determined based on the resource content features of the resource category. Then, the content form of the initial multimedia resource is expanded based on the content form features of the matched target resource category to obtain the corresponding target multimedia resource. That is, by first matching the resource content and then expanding the content form, the corresponding multimedia resource can be generated based on the resource features of the resource category, thereby improving the matching degree between the generated multimedia resource and each resource category.
[0060] In this embodiment, the step of expanding the content format of the initial multimedia resource based on the content format characteristics of the target resource classification to obtain the target multimedia resource includes:
[0061] When the content form characteristics of the target resource classification are inconsistent with the content form of the initial multimedia resource, the content form of the initial multimedia resource is transformed to obtain the target multimedia resource.
[0062] If the content format characteristics of the target resource category are consistent with the content format of the initial multimedia resource, the initial multimedia resource is expanded to have the same content format to obtain the target multimedia resource.
[0063] When the content form characteristics of the target resource category are inconsistent with the content form of the initial multimedia resource, for example, the content form of the initial multimedia resource is a first content form and the content form of the target resource category is a second content form, the content form of the initial multimedia resource can be transformed accordingly to obtain the target multimedia resource with the second content form.
[0064] When the content format characteristics of the target resource category are consistent with the content format of the initial multimedia resource, for example, if the content format of the initial multimedia resource is a first content format and the content format of the target resource category is a first content format, the initial multimedia resource can be expanded with the same content format to obtain the target multimedia resource with the first content format.
[0065] When the initial multimedia resources include resource content in a first content form and resource content in a second content form, and the target resource category has the content form of the first content form, the initial multimedia resources in the first content form can be expanded using the same content form, and the initial multimedia resources in the second content form can be expanded using the same content form, to obtain the target multimedia resources in the first content form. The first content form can be text, and the second content form can be an image.
[0066] The transformation and expansion between different content formats can include transforming the content format from text to image, or transforming the content format from image to text; the expansion between the same formats can include text rewriting, image-to-image generation, etc., where text rewriting can be the process of rewriting text content in the form of text to obtain text content in the form of text, and image-to-image generation can be the process of generating image content based on content in the form of image.
[0067] Specifically, for example, the initial multimedia resource content is related to food 1, and the content format of the initial multimedia resource is text. Correspondingly, the health resource category may involve topics related to food 1, such as an introduction to food 1, healthy cooking methods for food 1, etc. The content format of the health resource category may include text and image formats. Accordingly, the initial multimedia resource in text format can be expanded to obtain a target multimedia resource in text format, and the initial multimedia resource in text format can be transformed to obtain a target multimedia resource in image format. Similarly, the wallpaper resource category may involve images containing food 1, and its content format is image format. Correspondingly, the initial multimedia resource in text format can be transformed to obtain a target multimedia resource in image format.
[0068] In this embodiment, the specific content format expansion method can be determined based on the content format characteristics of the target resource category and the content format of the initial multimedia resource. This can include expansion between different content formats and expansion between the same content format. This improves the efficiency of generating multimedia resources for the target resource category, increases the matching degree between multimedia resources and their corresponding resource categories, and enhances the diversity of multimedia resource content formats.
[0069] In this embodiment, during the generation of the target multimedia resource, the target object contained in the target multimedia resource can also be specified, and accordingly, an image of the target object containing the target object needs to be provided as a reference guide; please refer to [link to relevant documentation] for details. Figure 4 It illustrates a method for generating target multimedia resources based on an image adaptation model, which may include:
[0070] S410. When the content of the initial multimedia resource is text, the content of the target multimedia resource to be generated is an image, and the target multimedia resource contains a target object, obtain the target object image containing the target object and the image adaptation model; the image adaptation model includes an image feature extraction layer, a first cross-attention layer and a second cross-attention layer.
[0071] This embodiment employs a method based on cue learning and a base model to generate multimedia resources. The cue learning process utilizes cue text and cue images. The base model can specifically be a model for content format expansion, including text-to-image models, image-to-text models, text rewriting models, and image-to-image models. The target object image is an image containing image information of the target object, and its image features can be extracted.
[0072] The image adaptation model may include an image feature extraction layer for extracting image features of the target object image, a first cross attention layer for performing feature cross processing on the image features, and a second cross attention layer for performing feature cross processing on the text features. The first cross attention layer and the second cross attention layer are decoupled from each other.
[0073] S420. Based on the image feature extraction layer, perform image feature extraction on the target object image to obtain the first image feature corresponding to the target object.
[0074] The first image feature can be an image feature extracted directly from the image of the target object.
[0075] S430. Based on the first cross-attention layer, perform feature cross-processing on the first image features to obtain the second image features.
[0076] The second image feature can be the image feature obtained by performing feature cross processing on the first image feature based on an attention mechanism.
[0077] S440. Extract text features from the initial multimedia resource to obtain the first text feature corresponding to the initial multimedia resource.
[0078] The first text feature can be a text feature extracted directly from the initial multimedia resource.
[0079] S450. Based on the second cross-attention layer, perform feature cross-processing on the first text features to obtain the second text features.
[0080] The second text feature can be the text feature obtained by performing feature cross processing on the first text feature based on an attention mechanism.
[0081] S460. The second image feature and the second text feature are respectively embedded into multiple diffusion processing layers in the text-to-image model, and the second image feature and the second text feature are processed based on the multiple diffusion processing layers to obtain a target multimedia resource whose content is an image and contains the target object.
[0082] In each diffusion processing layer, the second image features and the second text features are processed separately to obtain the output of each diffusion processing layer. Specifically, the text-to-image model can be a model capable of generating images based on text prompts. In this embodiment, since a target object image is introduced, which serves as an image prompt, it achieves the generation of target multimedia resources containing the target object in image form based on text and image prompts.
[0083] For example, if the text prompt is "Generate a person's smiling face", the image prompt can be an image containing object 1, and the corresponding generated target multimedia resource is an image containing the smiling face of object 1.
[0084] Please see Figure 5 The diagram illustrates the joint structure of the image adaptation model and the text-to-image model. The image adaptation model includes an image feature extraction layer, a first cross-attention layer, and a second cross-attention layer, which are decoupled. The text-to-image model may include a text feature extraction layer and a diffusion model, where the diffusion model may include multiple diffusion processing layers. The second image feature output from the first cross-attention layer can be used as the input to each diffusion processing layer, and the second text feature output from the second cross-attention layer can be used as the input to each diffusion processing layer. Each diffusion processing layer can process the second image feature and the second text feature separately, thus obtaining the output of each diffusion processing layer. Finally, the output of the text-to-image model is obtained, which is the target multimedia resource.
[0085] The image adaptation model in this embodiment can be used in a pre-trained text-to-image model to enable the generation of corresponding images using image prompts. The feature intersections of text features and image features are decoupled, meaning that text features and image features are processed separately. This fully utilizes the feature information of each feature during image generation, thereby improving the accuracy of image generation based on image prompts.
[0086] Furthermore, in this embodiment, publishing the target multimedia resource under the target resource category includes:
[0087] Create platform publishing accounts corresponding to each of the multiple resource categories, and configure the account publishing configuration information corresponding to the multiple platform publishing accounts;
[0088] Based on the platform publishing account corresponding to the target resource category and the account publishing configuration information corresponding to the platform publishing account corresponding to the target resource category, the target multimedia resource is published to the target resource category.
[0089] Each resource category on the target publishing platform can correspond to a platform publishing account. The account publishing configuration information characterizes the resource publishing characteristics of the corresponding platform publishing account, indicating the resource publishing time, such as the timing and frequency of account publishing. The timing of account publishing can represent the specific point in time when the account publishes multimedia resources, or it can be determined based on the timing of the target event. The frequency of account publishing can represent how often the account publishes multimedia resources, such as once per hour or once per minute. Each platform publishing account can publish multimedia resources to its corresponding resource category at the current moment, provided that the account publishing configuration information is met.
[0090] In this embodiment, by creating a corresponding platform publishing account for each resource category and configuring corresponding account publishing configuration information for each platform publishing account, the multimedia resources can be published to the corresponding resource category when the account publishing configuration information is met. This increases the verticality of the platform publishing accounts and improves the accuracy of resource publishing.
[0091] As can be seen from the above, there is a correspondence between the resource classification and the platform publishing account in this embodiment, and there can also be a correspondence between the corresponding platform publishing account and the multimedia resources to be published, so as to determine the multimedia resources to be published by the platform publishing account; please refer to the following for details. Figure 6 It illustrates a resource publishing method, which may include:
[0092] S610. Create an association between the platform publishing account corresponding to the target resource category and the target multimedia resource.
[0093] Each target resource category can have a corresponding platform publishing account, and the generated target multimedia resources also have a corresponding target resource category. This allows for the establishment of an association between the platform publishing account of the target resource category and the target multimedia resource. Specifically, it can establish an association between the account identifier of the platform publishing account and the resource identifier of the target multimedia resource. For example, target resource category 1 corresponds to platform publishing account 1, and target resource category 1 has multimedia resource 1 and multimedia resource 2 to be published. Accordingly, an association can be established between platform publishing account 1 and multimedia resource 1, and also between platform publishing account 1 and multimedia resource 2.
[0094] S620. If the account publishing configuration information of the platform publishing account corresponding to the target resource category is satisfied at the current moment, determine the multimedia resources to be published that are associated with the platform publishing account corresponding to the target resource category based on the association relationship; the multimedia resources to be published include the target multimedia resources.
[0095] If the platform publishing account's publishing configuration information for the target resource category is met at the current moment, it can be determined that the target multimedia resources to be published under the target resource category need to be published. This can be achieved by first identifying the corresponding multimedia resources to be published based on the association between the platform publishing account and the target multimedia resources. Since each multimedia resource publication involves publishing multimedia resources generated within a time period prior to the current moment, this may include other multimedia resources besides the target multimedia resources. These other multimedia resources could be generated from different initial multimedia resources or obtained from different channels; therefore, the multimedia resources to be published can include both the target multimedia resources and other multimedia resources.
[0096] S630. Publish the multimedia resource to be published to the target resource category based on the platform publishing account corresponding to the target resource category.
[0097] When there are multiple multimedia resources to be published, multiple multimedia resources can be published in batches to the target resource category through the platform publishing account corresponding to the target resource category.
[0098] For example, target resource category 1 corresponds to platform publishing account 1. Target resource category 1 has multimedia resource 1 and multimedia resource 2 to be published. Accordingly, a relationship can be established between platform publishing account 1 and multimedia resource 1, and a relationship can be established between platform publishing account 1 and multimedia resource 2. If the account publishing configuration information of platform publishing account 1 is met at the current time, multimedia resource 1 and multimedia resource 2 can be published in batches to target resource category 1 through platform publishing account 1.
[0099] Once the multimedia resources to be published corresponding to the platform publishing account are identified, the corresponding multimedia resources can be published based on the platform publishing account. This means that when resources need to be published, the multimedia resources to be published associated with the platform publishing account are first identified, and then the multimedia resources to be published are published in batches to the corresponding resource categories through the platform publishing account, which improves the efficiency and accuracy of resource publishing.
[0100] The implementation process of this application is illustrated below with a specific example. Please refer to [link / reference]. Figure 7 It illustrates a multimedia resource processing framework, which may include a creative source module, a multimedia resource generation module, a content review module, a content processing module, and a batch publishing module. The functions of each module are described below:
[0101] 1. Creative Source Module
[0102] The sources of creative content generation include: (a) popular content category mining: statistically analyzing the ranking of browsing content at all levels and related tags; (b) the topic tags selected or included in the published content, and then statistically analyzing the top topic tags corresponding to the relevant content; (c) the top search information of searched content; and (d) through a large language model, inputting relevant background and content scenarios to generate corresponding creative content themes. The creative themes obtained through the above methods are then aggregated, and a portion of the content is selected as creative themes based on business needs. In practical applications, this data can be used to construct a creative importance ranking model to assist in the selection process.
[0103] 2. Multimedia resource generation module
[0104] In this embodiment, the resource categories can be the channels within the target publishing platform. Channel post content includes materials such as images and text; images can include pictures and videos. The basic models for multimedia resource generation include the LLM (Large Language Model) and the MLLM (Multimodal Large Language Model). The LLM is primarily used for text content generation and processing, while the MLLM is used for understanding image content and configuring text for images. The diffusion model is mainly used for text-guided image generation and image-to-image generation. Typically, these models are used as a foundation, and the basic language model or diffusion model is fine-tuned using SFT or LoRA methods. This ensures the generated content better meets the actual needs of the corresponding scenario. Some content also needs to include preference information such as aesthetics and character design, such as images for movies and TV shows, which may be related to a specific person. In this case, based on the diffusion model, an image adaptation model can be used to fine-tune the adaptation model to generate multimedia resources containing information about the specified person, making the final multimedia resources more suitable for the user's needs. For generating highly timely information, a search engine is needed to search for relevant topics as context for rewriting and inputting the LLM language model, thus obtaining the latest, most timely content.
[0105] 3. Content moderation module
[0106] Primarily to ensure that the generated multimedia resources meet content quality and security requirements, and to avoid some sensitive issues that are difficult to judge, security quality models and content quality models can be used to provide corresponding review capabilities.
[0107] 4. Content Processing Module
[0108] The final published multimedia resources are usually content posts, which need to be formatted and organized according to a certain layout. Content posts can be plain text, pure images (typically 6 or 9 images), or a combination of text and images (e.g., images with accompanying text). Publishing templates corresponding to various publishing formats can also be pre-created, such as plain text templates, pure image templates, and combined text and image templates. Publishing templates can include layout styles, template backgrounds, template fonts, and template styles.
[0109] 5. Batch Release Module
[0110] This process includes two steps: multimedia resource selection and multimedia resource export. Multimedia resource selection involves obtaining resource identifiers for multimedia resources, typically generated uniquely by the business system. Multimedia resource export mainly refers to the platform publishing account used for bulk content publishing, along with the account's publishing configuration information. This can be customized based on the specific needs of the business. It allows establishing a link between the platform publishing account's account identifier and the multimedia resource's resource identifier, enabling the publishing of associated multimedia resources to the corresponding channels based on the platform publishing account.
[0111] This embodiment also provides a method and system for automated creation and generation of channel content based on a generative large model. The main functions of each service module are as follows:
[0112] I. End
[0113] (1) By communicating with the message and content service access server, the channel completes various community functions such as uploading pictures, publishing videos and exchanging text content. In addition, PGC or UGC, MCN or PUGC content producers provide local or shot videos through mobile terminals or backend API systems. These are the main sources of content distribution, which can also be considered as a broad end. The automated creation and generation auxiliary pipeline is another content supply end based on AIGC.
[0114] (2) The carrier of various functions in the content and social business ecosystem, such as chat, live streaming, creation, etc. in the channel;
[0115] (3) When publishing content, you usually first obtain the upload server interface address, and then upload the local file. During the shooting process, you can choose to match the local text and image content with music, filter templates and filter beautification functions, etc.
[0116] (4) Communicate with the reporting and analysis interface server to collect detailed data in various sub-business scenarios under the social network scenario, and save the collected data in the statistical analysis database as the basic data source and creative source for building and training the channel content automated creation and generation model;
[0117] II. Content Service Access Server
[0118] (1) Synchronize with the terminal to complete the uplink and downlink communication and synchronization of various information in the channel community function;
[0119] (2) Communicate directly with the content production end. The content submitted from the front end is usually the title, publisher, summary, cover image, and publication time of the content, or the video taken is directly sent to the server and stored in the channel content database.
[0120] (3) Write the video content's metadata, such as file size, cover image link, bitrate, file format, title, release time, author, etc., into the channel content database;
[0121] III. Content Database
[0122] (1) Temporarily save user-posted content to enable content roaming and multi-terminal message synchronization;
[0123] (2) As the core module of the community system, it efficiently optimizes the storage and indexing of channel content and provides a source of information that is synchronized across multiple devices;
[0124] (3) The core database of content: all the metadata of the content published by producers is stored in this business database. The focus is on the metadata of the content itself, such as image file size, image link, quantity, name and necessary description, file format, title, publication time, author, whether it is original, source (such as UGC, MCN or AIGC generated), etc.
[0125] IV. Business Systems
[0126] (1) Responsible for the synchronization, circulation, scheduling, distribution, and updating of community content, as well as the maintenance of content status;
[0127] (2) Responsible for communicating with the content database and completing the distribution and processing of relevant content, such as channel search, channel post search and recommendation, etc.;
[0128] V. Reporting and Analysis Interface Service
[0129] (1) Communicate with the end to receive various feedback during the message consumption and distribution process, such as feedback on the quality of content distribution;
[0130] (2) The terminal reports data in different business sub-scenarios. After real-time data cleaning, the data is stored in different storage engines. Combined with the content flow of different business scenarios, the data and feedback information required for training the upper-layer basic model are constructed.
[0131] VI. Statistical Analysis Database
[0132] (1) And the reporting and analysis interface service, save the channel content after desensitization and perform preliminary processing of data cleaning and verification for different business scenarios;
[0133] (2) For community content, interactive and feedback information will be selected and reported to facilitate fine-tuning of the model and evaluation of the actual model understanding and generation effect for subsequent improvement and optimization;
[0134] VII. Content Automated Production Processing Pipeline
[0135] (1) As described above, the generative capabilities provided by the generative big model are called on the basis of creativity. The big model here includes text-to-text model, image-to-text model and image-to-image model, etc., as well as quality-related models. Typical content generation and creation paths include: learning through prompts + powerful base models, expanding text content based on existing images to generate high-quality channel post titles and copywriting. Another generation path is: theme-based cultural and creative content is mainly based on platform hot topics or personalized theme content of specific groups. After generating copywriting materials, images and other modal content are expanded, that is, text-to-image, etc., as well as image-to-image and copywriting polishing and rewriting. Finally, the combination of images and text is used to obtain the final distribution of various types of posts.
[0136] (2) In addition to generating content, the pipeline also needs to schedule content review, content processing and batch publishing related services, and maintain linkage with the channel digital content distribution system;
[0137] 8. Digital Content Distribution System
[0138] (1) Content is distributed through search and recommendation to accelerate the distribution of content across various channels, while maintaining communication with sub-business systems and automated production pipelines;
[0139] IX. Sub-business Services
[0140] (1) Usually, it is the platform's business subsystem, such as publishing text, publishing pictures, content interaction, and other business subsystems and services, such as search channels, creating channels, and various management tools within the channels;
[0141] 10. Instruction Fine-tuning Sample Library and Creative Theme Library
[0142] (1) Mainly, it is necessary to prepare the instruction fine-tuning samples and through the basic model for different creative content styles and styles required by different channels and sub-channels. Fine-tuning methods include SFT, LoRA, and if personalized content is generated, an image adaptation model will be used to adapt the image to the person.
[0143] (2) Creative theme library, the source of content generation. This mainly generates vertical channel content based on platform hot words and platform top search queries, which can be used to supplement the platform's content ecosystem, improve user stickiness, and effectively reduce the cost of content supply and increase content supply.
[0144] XI. Generative Large Model Groups
[0145] (1) This is not limited to a fixed large language model. Any model that uses a large amount of rich basic corpus to construct a large-scale generative Transform architecture can be classified into this category. For example, in the specific implementation, the language model can be based on LLaMa and GLM models. Visual modal support is added on the basis of the basic language model, which enables the model to have good image perception and understanding capabilities and has also achieved rapid development. These models can be better adapted to business scenarios and needs after being fine-tuned by SFT and LoRA through scene-related data samples. Text-to-text and image-to-text use basic language models and multimodal large models.
[0146] This paper employs a diffusion model as the foundational model for text-to-image and image-to-image generation. It's an advanced image generation method that significantly lowers the barrier to entry. The diffusion model works based on the diffusion process, which transforms data from a high-dimensional distribution to a low-dimensional distribution. In image generation, images can be viewed as high-dimensional data, and the diffusion model transforms this high-dimensional data into low-dimensional data by adding randomness at each step, thus generating new images. The core of the model is a U-Net network structure, which, through training, learns to predict the added noise after inputting a noisy image. Thus, during prediction, when given a blurred image, the U-Net removes the predicted noise, reconstructing the original image as closely as possible. Based on this, LoRA is used for fine-tuning, while direct text guidance is employed to generate the required content.
[0147] The multimedia resource processing method in this embodiment can fully utilize post-hoc distribution data and obtain creative sources for multimedia resource production from multiple channels. For example, it can mine creative intent based on top multimedia resources viewed and search information from different themes in the channel community, serving as an effective source of creative ideas for multimedia resource generation. This ensures that the generated multimedia resources can well meet the actual needs of users and the needs of the platform, effectively reducing the threshold for the cold start of community multimedia resources, significantly reducing the manual collection and operation by operators, and effectively reducing the labor costs of multimedia resource operation. It can automatically produce and publish multimedia resources in batches, meeting the timeliness requirements of multimedia resource production. It can optimize the degree of automation and robustness, thereby obtaining a better user multimedia resource experience, and ultimately accelerating the activation of social relationship chains and improving the overall activity of the platform.
[0148] It should be noted that any of the methods described above in this embodiment can be combined based on the actual implementation situation and have corresponding beneficial effects, which will not be elaborated here.
[0149] This embodiment also provides a multimedia resource processing device; please refer to [link / reference]. Figure 8 The device may include:
[0150] Resource acquisition module 810 is used to acquire initial multimedia resources;
[0151] The resource processing module 820 is used to expand the multimedia resource form of the initial multimedia resource based on the association information between the initial multimedia resource and multiple resource categories in the target publishing platform, so as to obtain a target multimedia resource corresponding to the target resource category; the target resource category is the resource category associated with the initial multimedia resource among the multiple resource categories, and the multimedia resource form of the target multimedia resource matches the target resource category;
[0152] The resource publishing module 830 is used to publish the target multimedia resource under the target resource category.
[0153] Furthermore, the resource processing module includes:
[0154] The resource feature acquisition module is used to acquire the resource features corresponding to each of the multiple resource categories; the resource features corresponding to each resource category include multimedia resource features and multimedia resource form features.
[0155] A matching module is used to match the multimedia resource features of the multiple resource categories with the multimedia resources of the initial multimedia resource, and determine the target resource category from the multiple resource categories based on the multimedia resource matching results; the multimedia resource features of the target resource category are matched with the multimedia resources of the initial multimedia resource.
[0156] The morphology expansion module is used to expand the initial multimedia resource morphology based on the multimedia resource morphology characteristics of the target resource classification to obtain the target multimedia resource.
[0157] Furthermore, the morphology extension module is used for:
[0158] When the multimedia resource form characteristics of the target resource classification are inconsistent with the multimedia resource form of the initial multimedia resource, the multimedia resource form of the initial multimedia resource is transformed to obtain the target multimedia resource.
[0159] If the multimedia resource format characteristics of the target resource category are consistent with the multimedia resource format of the initial multimedia resource, the initial multimedia resource is expanded to have the same multimedia resource format to obtain the target multimedia resource.
[0160] Furthermore, the step of expanding the initial multimedia resource format to obtain the target multimedia resource corresponding to the target resource category includes:
[0161] If the initial multimedia resource is in text format, the target multimedia resource to be generated is in image format, and the target multimedia resource contains a target object, then the target object image containing the target object and the image adaptation model are obtained; the image adaptation model includes an image feature extraction layer, a first cross-attention layer, and a second cross-attention layer.
[0162] Based on the image feature extraction layer, image features are extracted from the target object image to obtain the first image feature corresponding to the target object;
[0163] Based on the first cross-attention layer, feature cross processing is performed on the first image features to obtain the second image features;
[0164] Text features are extracted from the initial multimedia resources to obtain the first text features corresponding to the initial multimedia resources;
[0165] Based on the second cross-attention layer, the first text features are subjected to feature cross processing to obtain the second text features;
[0166] The second image feature and the second text feature are respectively embedded into multiple diffusion processing layers in the text-to-image model. The second image feature and the second text feature are processed based on the multiple diffusion processing layers to obtain a target multimedia resource in the form of an image that contains the target object.
[0167] Furthermore, the resource acquisition module is used for:
[0168] Obtain historical multimedia resources published by the user account, resource search information of the user account, and popular multimedia resources;
[0169] Based on the historical multimedia resources, the resource search information, and the hot multimedia resources, resource analysis is performed to obtain the initial multimedia resources.
[0170] Furthermore, the resource publishing module includes:
[0171] The account creation module is used to create platform publishing accounts corresponding to the multiple resource categories, and to configure the account publishing configuration information corresponding to the multiple platform publishing accounts;
[0172] The associated publishing module is used to publish the target multimedia resource to the target resource category based on the platform publishing account corresponding to the target resource category and the account publishing configuration information corresponding to the platform publishing account corresponding to the target resource category.
[0173] Furthermore, the associated publishing module is used for:
[0174] Create an association between the platform publishing account corresponding to the target resource category and the target multimedia resource;
[0175] If the account publishing configuration information of the platform publishing account corresponding to the target resource category is satisfied at the current moment, the multimedia resources to be published associated with the platform publishing account corresponding to the target resource category are determined based on the association relationship; the multimedia resources to be published include the target multimedia resources;
[0176] The multimedia resources to be published will be published to the target resource category based on the platform publishing account corresponding to the target resource category.
[0177] The apparatus provided in the above embodiments can execute the methods provided in any embodiment of this application, and has the corresponding functional modules and beneficial effects for executing the method. Technical details not described in detail in the above embodiments can be found in the methods provided in any embodiment of this application.
[0178] This embodiment also provides a computer-readable storage medium storing at least one instruction or at least one program, which is loaded by a processor and executed as any of the methods described above in this embodiment.
[0179] According to one aspect of this application, a computer program product or computer program is provided, comprising computer instructions stored in a computer-readable storage medium. A processor of an electronic device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform any of the methods described above.
[0180] Figure 9 This is a block diagram illustrating an electronic device for multimedia resource processing according to an exemplary embodiment. The electronic device may be a server, and its internal structure diagram may be as follows: Figure 9 As shown, this electronic device includes a processor, memory, and a network interface connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage medium. The network interface is used to communicate with external terminals via a network connection. When the computer program is executed by the processor, it implements a multimedia resource processing method.
[0181] Those skilled in the art will understand that Figure 9The structure shown is merely a block diagram of a portion of the structure related to the present disclosure and does not constitute a limitation on the electronic device to which the present disclosure is applied. A specific electronic device may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.
[0182] This specification provides the operational steps of the methods described in the embodiments or flowcharts, but more or fewer operational steps may be included based on conventional or non-inventive labor. The steps and order listed in the embodiments are merely one possible execution order among many steps and do not represent the only execution order. In actual system or interrupt product execution, the methods shown in the embodiments or drawings can be executed sequentially or in parallel (e.g., in a parallel processor or multi-threaded processing environment).
[0183] The structure shown in this embodiment is only a partial structure related to the solution of this application and does not constitute a limitation on the device to which the solution of this application is applied. Specific devices may include more or fewer components than shown, or combinations of certain components, or arrangements of different components. It should be understood that the methods, apparatuses, etc., disclosed in this embodiment can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For example, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection between devices or unit modules through some interfaces.
[0184] Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.
[0185] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed in this specification can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of each example have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.
[0186] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.
Claims
1. A method of processing multimedia resources, characterized in that, include: Obtain initial multimedia resources; Based on the association information between the resource content of the initial multimedia resource and multiple resource categories in the target publishing platform, the content format of the initial multimedia resource is expanded to obtain a target multimedia resource corresponding to the target resource category; the target resource category is the resource category among the multiple resource categories that is associated with the resource content of the initial multimedia resource, and the content format of the target multimedia resource matches the target resource category; The target multimedia resource is published under the target resource category.
2. The method of claim 1, wherein, The method involves expanding the content format of the initial multimedia resource based on the association information between the resource content and multiple resource categories in the target publishing platform, to obtain the target multimedia resource corresponding to the target resource category, including: Obtain the resource features corresponding to each of the multiple resource categories; the resource features corresponding to each resource category include resource content features and content form features; The resource content features of the multiple resource categories are matched with the resource content of the initial multimedia resource, and the target resource category is determined from the multiple resource categories based on the content matching results; the resource content features of the target resource category are matched with the resource content of the initial multimedia resource. Based on the content format characteristics of the target resource classification, the initial multimedia resource is expanded in content format to obtain the target multimedia resource.
3. The method of claim 2, wherein, The content format expansion of the initial multimedia resource based on the content format features of the target resource classification to obtain the target multimedia resource includes: When the content form characteristics of the target resource classification are inconsistent with the content form of the initial multimedia resource, the content form of the initial multimedia resource is transformed to obtain the target multimedia resource. If the content format characteristics of the target resource category are consistent with the content format of the initial multimedia resource, the initial multimedia resource is expanded to have the same content format to obtain the target multimedia resource.
4. The method of claim 1, wherein, The step of expanding the content format of the initial multimedia resources to obtain target multimedia resources corresponding to the target resource category includes: If the content of the initial multimedia resource is text, the content of the target multimedia resource to be generated is an image, and the target multimedia resource contains a target object, then the target object image containing the target object and the image adaptation model are obtained; the image adaptation model includes an image feature extraction layer, a first cross-attention layer and a second cross-attention layer. Based on the image feature extraction layer, image features are extracted from the target object image to obtain the first image feature corresponding to the target object; Based on the first cross-attention layer, feature cross processing is performed on the first image features to obtain the second image features; Text features are extracted from the initial multimedia resources to obtain the first text features corresponding to the initial multimedia resources; Based on the second cross-attention layer, the first text features are subjected to feature cross processing to obtain the second text features; The second image feature and the second text feature are respectively embedded into multiple diffusion processing layers in the text-to-image model. The second image feature and the second text feature are processed based on the multiple diffusion processing layers to obtain a target multimedia resource whose content is an image and contains the target object.
5. The method of claim 1, wherein, The acquisition of the initial multimedia resources includes: Obtain historical multimedia resources published by the user account, resource search information of the user account, and popular multimedia resources; Based on the historical multimedia resources, the resource search information, and the hot multimedia resources, resource analysis is performed to obtain the initial multimedia resources.
6. The method of claim 1, wherein, Publishing the target multimedia resource under the target resource category includes: Create platform publishing accounts corresponding to each of the multiple resource categories, and configure the account publishing configuration information corresponding to the multiple platform publishing accounts; Based on the platform publishing account corresponding to the target resource category and the account publishing configuration information corresponding to the platform publishing account corresponding to the target resource category, the target multimedia resource is published to the target resource category.
7. The method of claim 6, wherein, The step of publishing the target multimedia resource to the target resource category based on the platform publishing account corresponding to the target resource category and the account publishing configuration information corresponding to the platform publishing account corresponding to the target resource category includes: Create an association between the platform publishing account corresponding to the target resource category and the target multimedia resource; If the account publishing configuration information of the platform publishing account corresponding to the target resource category is satisfied at the current moment, the multimedia resources to be published associated with the platform publishing account corresponding to the target resource category are determined based on the association relationship; the multimedia resources to be published include the target multimedia resources; The multimedia resources to be published will be published to the target resource category based on the platform publishing account corresponding to the target resource category.
8. A multimedia resource processing apparatus, characterized by comprising: include: The resource acquisition module is used to acquire initial multimedia resources; The resource processing module is used to expand the content format of the initial multimedia resource based on the resource content of the initial multimedia resource and the association information of multiple resource categories in the target publishing platform, so as to obtain a target multimedia resource corresponding to the target resource category; the target resource category is the resource category associated with the resource content of the initial multimedia resource among the multiple resource categories, and the content format of the target multimedia resource matches the target resource category; The resource publishing module is used to publish the target multimedia resources under the target resource category.
9. An electronic device, comprising: The device includes a processor and a memory, the memory storing at least one instruction or at least one program, the at least one instruction or the at least one program being loaded and executed by the processor to implement the multimedia resource processing method as described in any one of claims 1 to 7.
10. A computer storage medium, characterized in that, The storage medium stores at least one instruction or at least one program, which is loaded and executed by a processor according to any one of claims 1 to 7.