AI exchange platform for creative content industry

Through an AI exchange platform for the creative content industry, integrating AI tool components and service resources, and integrating resources using AI learning capabilities, it solves the problems of traditional AI applications in short supply and data security, and achieves fast and flexible creative content creation and Efficient video and image generation.

CN118940861APending Publication Date: 2024-11-12JIANGSU CUDATEC TECH CO LTD
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202411121800.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-15
Publication Date
2024-11-12

Smart Images

  • Figure CN118940861A_ABST
    Figure CN118940861A_ABST
Patent Text Reader

Abstract

The invention discloses a creative content industry-oriented AI exchange platform, which is characterized in that the AI capability is utilized, the existing resources are learned to form new AI service resources and AI tools, and the existing tools and resources are inherited and integrated with the AI service resources and the AI tools to form the AI exchange platform, so that the creative content generation quality and efficiency are improved; the system specifically comprises an exchange middle table, tool resources, service resources, a personalized multi-agent scheme library and a computing power base, wherein the exchange middle table comprises an AI capability library and a model library; the tool resources comprise existing tools, plug-ins, functions and AI tools; the service resources comprise material resources required by task processing, AI customized service resources and AI knowledge base resources; the personalized multi-agent scheme library comprises a set of task processing schemes obtained by AI capability calling, AI generation and combination; the computing power base is a cloud elastic computing pool.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to an AI cloud service macro model, and in particular to an AI exchange platform for the creative content industry. Background Art

[0002] The construction background of the AI ​​cloud service model is rooted in the current urgent need for high-performance and high-efficiency AI applications. With the widespread application of AI technology in various industries, traditional local deployment faces problems such as insufficient resources, limited computing power and data security, which restricts the expansion of AI applications and the implementation of actual application scenarios. The cloud service model came into being. By concentrating high-performance computing resources and various functional components running in the cloud, it provides users with powerful data processing, analysis and prediction capabilities, greatly reduces deployment and maintenance costs, and can achieve instant updates and flexible expansion.

[0003] However, building an AI cloud service model also faces challenges such as data privacy, cross-platform compatibility, and real-time requirements. It is necessary to comprehensively consider multiple factors such as hardware infrastructure, network bandwidth, and security assurance in order to achieve comprehensive support and continuous innovation for AI application needs in the creative content industry. Summary of the invention

[0004] In response to the above technical problems, the purpose of the present invention is to provide an AI exchange platform for the creative content industry, integrate the AI ​​tool components and AI service resources required by the digital content creative industry, provide users with fast and flexible creative content creation services, and at the same time, learn from user behavior through AI learning capabilities, expand and enrich tool functions and design resources, continuously iterate and update design tools and material resources, and improve tool call flexibility and resource diversity.

[0005] In order to achieve the above-mentioned purpose, the technical solution adopted by the present invention is: an AI exchange platform for the creative content industry, which utilizes AI capabilities to learn from existing resources such as historical data, user behavior, historical tasks, etc., to form new AI service resources and AI tools, and inherit and integrate existing tools and resources with AI service resources and AI tools to improve the quality and efficiency of creative content generation.

[0006] As a preferred embodiment of the present invention, the infrastructure of the AI ​​exchange platform includes an exchange middle platform, tool resources, service resources, a personalized multi-agent solution library and a computing power base. The exchange middle platform includes an AI capability library and a model library; tool resources include existing tools, plug-ins, functions and AI tools; service resources include material resources required for task processing, AI customized service resources and AI knowledge base resources; the personalized multi-agent solution library includes a collection of task processing solutions obtained by AI capability call, AI generation and combination; the computing power base is a cloud-based elastic computing pool.

[0007] As a preferred embodiment of the present invention, tool resources can be classified according to categories, including management capabilities, business capabilities and security capabilities. The management capabilities include sub-account management and background process monitoring, the business capabilities include automatic AI scene generation and AI rendering, and the security capabilities include data storage and data asset traceability.

[0008] As a preferred embodiment of the present invention, the material resources include local materials and AI-generated materials, the AI ​​customized service resources include customized model construction, personalized computing power streaming, customized exclusive host call and personalized preloading, and the AI ​​knowledge base resources include automatic material identification and retrieval, data learning and analysis, and resource iterative update.

[0009] As a preferred method of the present invention, when a user task is input into the exchange center, the exchange center first calls the corresponding AI capability in the AI ​​capability library, extracts the task features, matches the task processing solution in the personalized multi-agent solution library, and performs task processing;

[0010] If the personalized multi-agent solution library does not have a suitable task processing solution, the task processing model is matched in the model library, and the tool resources and service resources are interacted with. The adapted tool resources and service resources are called according to the category to generate a new personalized multi-agent solution, which is delivered to the user after the task processing is completed. At the same time, the solution increment formed is recorded in the solution library. When the same task requirement is encountered next time, the multi-agent solution is immediately called to quickly respond and process.

[0011] As a preferred embodiment of the present invention, the task characteristics include task requirements, usage scenarios and task stages.

[0012] As a preferred embodiment of the present invention, the task processing solution includes service resources and tool resources required for task processing.

[0013] As a preferred embodiment of the present invention, the AI ​​exchange platform has the ability to process tasks in parallel:

[0014] When a user inputs a task, the exchange middleware calls the corresponding AI capabilities in the AI ​​capability library to split the task, matches the task processing model in the model library, adapts resources and solutions through the gated network, selects appropriate tool resources and service resources according to the task stage, process and type, and dynamically allocates tasks to corresponding resources for parallel task processing. Resource allocation is performed according to the weight of the resources required for task processing. After the task processing is completed, the processing results are weighted added to form the final output result.

[0015] Compared with the prior art, the present invention has the following technical effects:

[0016] 1. The present invention integrates the AI ​​tool components and AI service resources required by the digital content creative industry to provide users with fast and flexible creative content creation services.

[0017] 2. The present invention uses AI learning capabilities to learn user behaviors, expand and enrich tool functions and service resources, continuously iterate and update design tools and material resources, and improve tool calling flexibility and resource diversity.

[0018] 3. For different task scenarios, since the tools and resources required for each stage are different, the degree to which AI can intervene in intelligent processing is also different. The present invention integrates the AI ​​tools and capabilities involved in various usage scenarios, which can realize one-click generation of videos and images, simplify the production process, and provide personalized design solutions. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 Schematic diagram of the infrastructure of the AI ​​exchange platform for the creative content industry in this embodiment.

[0020] Figure 2 Schematic diagram of capability exchange of the AI ​​exchange platform for the creative content industry in this embodiment.

[0021] Figure 3 Schematic diagram of task parallel processing of the AI ​​exchange platform for the creative content industry in this embodiment. DETAILED DESCRIPTION

[0022] In order to explain the technical content, structural features, achieved objectives and effects of the technical solution in detail, the following is a detailed description in conjunction with specific embodiments and accompanying drawings.

[0023] The AI ​​exchange platform described in this embodiment refers to the use of AI capabilities to learn from existing resources such as historical data, user behavior, and historical tasks to form new AI service resources and AI tools, and inherit and integrate existing tools and resources with AI resources to improve the quality and efficiency of creative content generation.

[0024] like Figure 1 As shown, the AI ​​exchange platform infrastructure for the creative content industry:

[0025] It includes an exchange platform, capability resources, service resources, and a personalized multi-agent solution library (referred to as the solution library). The exchange platform includes an AI capability library and a model library; capability resources include existing tools, plug-ins, functions, etc., and tool resources are classified according to categories, such as management capabilities (parent-child account management, background process monitoring, etc.), business capabilities (automatic generation of AI scenes, AI rendering), and security capabilities (data storage, data asset traceability), etc.; service resources include material resources required for task processing (local materials, AI-generated materials, etc.), AI customized service resources (customized model building, personalized computing power streaming, customized exclusive host calls, personalized preloading), and AI knowledge base resources (automatic material identification and retrieval, data learning and analysis, and resource iterative updates); the personalized multi-agent solution library includes a collection of task processing solutions obtained by calling, generating, and combining AI capabilities; the computing power base is the cloud-based elastic computing pool.

[0026] When the user's task is input into the exchange center, the exchange center first calls the corresponding AI capability in the AI ​​capability library, extracts the task requirements, usage scenarios, task stages and other features, matches the task processing solution in the solution library, and performs task processing. The task processing solution includes the service resources and capability resources required for task processing.

[0027] If there is no suitable task processing solution in the solution library, the task processing model will be matched in the model library, and the capability resources and service resources will be interacted with to call the most suitable capability resources and service resources by category to generate a new personalized multi-agent solution, which will be delivered to the user after the task processing is completed. At the same time, the solution increment formed will be recorded in the solution library. The next time the same task requirement is encountered, the multi-agent solution will be called immediately for rapid response and processing.

[0028] like Figure 2 As shown in the figure, for task scenarios such as image production, video generation, home decoration design, and product design, the degree to which AI can intervene in intelligent processing varies due to the different tools and resources required at each stage. By integrating the AI ​​tools and capabilities involved in each usage scenario through the exchange platform, one-click generation of videos and images can be achieved, the production process can be simplified, and personalized design solutions can be provided. By perceiving the scene type, thinking about the AI ​​tools and capabilities required at different stages of the learning task, and calling the corresponding task processing solution to complete the creation of digital content, a specific effect can be achieved.

[0029] like Figure 3 As shown, a schematic diagram of task parallel processing of the AI ​​exchange platform for the creative content industry in this embodiment.

[0030] When a task is input, the switching center exchanges the corresponding AI capabilities in the AI ​​capability library to split the task, matches the task processing model in the model library, adapts resources and solutions through the gated network, selects appropriate tool resources and service resources according to the task stage, process, and type, and dynamically allocates tasks to corresponding resources for parallel task processing. It allocates resources according to the weight of the resources required for task processing, and performs weighted addition of the processing results after the task processing is completed.

[0031] For example, in the rendering design stage, the image quality conversion, auxiliary generation and other functions in the AI ​​service resources are mainly used, while AI rendering is less used in this process. However, in the rendering stage, the AI ​​rendering function is used more. Therefore, resources are allocated to tasks in different stages according to weights, and finally the results are weighted added to form the final output result, thereby improving resource utilization efficiency and task processing accuracy.

[0032] In this embodiment, the exchange platform processes tasks simultaneously through multiple nodes or servers, splits tasks into independently complete subtasks and assigns them to multiple computing nodes. Each node processes a part of the task processing work, and uses distributed task processing to improve resource utilization and task processing efficiency. At the same time, the user's task processing status is monitored in real time, and rendering tasks are dynamically assigned based on computing resources and task requirements. Through the user feedback mechanism, user usage behaviors and usage problems are collected, and the resource library is continuously optimized, iterated, and enriched.

[0033] It should be noted that, although the above embodiments have been described in this article, the patent protection scope of the present invention is not limited thereby. Therefore, based on the innovative concept of the present invention, changes and modifications made to the embodiments described herein, or equivalent structures or equivalent process changes made using the contents of the present invention specification and drawings, directly or indirectly applying the above technical solutions to other related technical fields, are all included in the patent protection scope of the present invention.

Claims

1. An AI exchange platform for the creative content industry, characterized by: Utilize AI capabilities and learn from existing resources to form new AI service resources and AI tools, and inherit and integrate existing tools and resources with AI service resources and AI tools to form an AI exchange platform.

2. The AI ​​exchange platform for the creative content industry according to claim 1, characterized in that: It includes an exchange middle platform, tool resources, service resources, a personalized multi-agent solution library and a computing power base. The exchange middle platform includes an AI capability library and a model library; tool resources include existing tools, plug-ins, functions and AI tools; service resources include the material resources required for task processing, AI customized service resources and AI knowledge base resources; the personalized multi-agent solution library includes a collection of task processing solutions obtained by calling, generating and combining AI capabilities; the computing power base is the cloud-based elastic computing pool.

3. The AI ​​exchange platform for the creative content industry according to claim 2, characterized in that: Tool resources are classified according to categories, including management capabilities, business capabilities and security capabilities. The management capabilities include parent-child account management and background process monitoring, the business capabilities include automatic AI scene generation and AI rendering, and the security capabilities include data storage and data asset traceability.

4. The AI ​​exchange platform for the creative content industry according to claim 2, characterized in that: The material resources include local materials and AI-generated materials, the AI ​​customized service resources include customized model construction, personalized computing power streaming, customized exclusive host call and personalized preloading, and the AI ​​knowledge base resources include automatic material identification and retrieval, data learning and analysis, and resource iterative updates.

5. The AI ​​exchange platform for the creative content industry according to claim 2, characterized in that: When the user's task is input into the exchange center, the exchange center first calls the corresponding AI capability in the AI ​​capability library, extracts the task features, matches the task processing solution in the personalized multi-agent solution library, and performs task processing; If the personalized multi-agent solution library does not have a suitable task processing solution, the task processing model is matched in the model library, and the tool resources and service resources are interacted with. The adapted tool resources and service resources are called according to the category to generate a new personalized multi-agent solution, which is delivered to the user after the task processing is completed. At the same time, the solution increment formed is recorded in the solution library. When the same task requirement is encountered next time, the multi-agent solution is immediately called to quickly respond and process.

6. The AI ​​exchange platform for the creative content industry according to claim 5, characterized in that: The task characteristics include task requirements, usage scenarios and task stages.

7. The AI ​​exchange platform for the creative content industry according to claim 5, characterized in that: The task processing solution includes service resources and tool resources required for task processing.

8. The AI ​​exchange platform for the creative content industry according to claim 2, characterized in that: The AI ​​switching platform has the ability to process tasks in parallel: When a user inputs a task, the exchange middleware calls the corresponding AI capabilities in the AI ​​capability library to split the task, matches the task processing model in the model library, adapts resources and solutions through the gated network, selects appropriate tool resources and service resources according to the task stage, process and type, and dynamically allocates tasks to corresponding resources for parallel task processing. Resource allocation is performed according to the weight of the resources required for task processing. After the task processing is completed, the processing results are weighted added to form the final output result.

Citation Information

Cited By

  • AI workflow automatic management method based on creative scene

    CN119248458A