An intelligent video creation system supporting UGC community, scene-to-scene piece, and virtual live broadcast

Abstract

The application discloses an intelligent video creation system supporting UGC community, scene-to-picture transition and virtual live broadcast, comprising a front-end display interaction module for providing template selection, material uploading, video generation and preview functions, an industry template library module for storing industry customized template data derived from real service cases, a one-key picture making module for automatically completing video editing, music matching, letter generation and scene-to-picture transition based on an AI one-key picture making engine according to the template and the uploaded material selected by a user, a virtual live broadcast assistant module for real-time fusion and output of a person and a virtual background through AI image extraction technology, a creator community module for user uploading, sharing, evaluation and transaction of original templates, formation of an UGC content ecology and a training and teaching module for providing short video creation teaching courses, practical operation and skill authentication functions for the user; the application reduces the video creation threshold, improves the content production efficiency and builds a sustainable content ecology through the template, intelligence and community.

Description

Technical Field

[0001] This invention relates to the field of software as a service technology, and in particular to an intelligent video creation system that supports UGC communities, transitional video production, and virtual live streaming. Background Technology

[0002] With the rise of short video content marketing, more and more small and medium-sized enterprises and individual creators need efficient and low-cost video production tools.

[0003] However, traditional video production tools are complex to operate. Video production, live streaming tools, teaching systems, and community ecosystems are independent of each other. Users need to switch between different platforms. The functional modules are fragmented, resulting in high learning costs. At the same time, current one-click video production technology lacks template support for specific industries (such as clothing, jewelry, food, etc.). Template updates depend on the platform's R&D team, resulting in slow iteration and difficulty in meeting the rapidly changing market demands. Furthermore, its video editing logic is mostly fixed and cannot intelligently adjust the editing rhythm, music mood, and transition methods according to the content of the material. Existing software as a service systems mostly focus on general functions, which makes it difficult to meet the refined needs of the consumer goods industry for visual style, narrative structure and industry characteristics. Therefore, this invention proposes an intelligent video creation system that supports UGC communities, transitions, and virtual live streaming to solve the problems existing in the prior art. Summary of the Invention

[0004] To address the aforementioned issues, the present invention aims to propose an intelligent video creation system that supports UGC communities, transitions, and virtual live streaming. This intelligent video creation system, through template-based, intelligent, and community-based approaches, lowers the barrier to video creation, improves content production efficiency, and builds a sustainable content ecosystem. It achieves full automation from uploading materials to outputting the final product, significantly reducing the technical threshold and time cost of short video production.

[0005] To achieve the objectives of this invention, the following technical solution is provided: an intelligent video creation system supporting UGC communities, transition-based video production, and virtual live streaming, comprising a front-end display and interaction module integrated into a WeChat mini-program, an industry template library module, a one-click video production module, a virtual live streaming assistant module, a creator community module, and a training and teaching module. The front-end display and interaction module provides template selection, material upload, video generation, and preview functions. The industry template library module stores industry-customized template data derived from real service cases. The one-click video production module, based on an AI one-click video production engine, automatically completes video editing, music addition, letter generation, and transition synthesis according to the user-selected template and uploaded materials. The virtual live streaming assistant module uses AI keying technology to perform real-time fusion output of characters and virtual backgrounds. The creator community module allows users to upload, share, evaluate, and trade original templates, forming a UGC content ecosystem. The training and teaching module provides users with short video creation courses, practical exercises, and skills certification functions.

[0006] Further improvements include: the front-end display and interaction module includes template category browsing, intelligent search, one-click material upload, real-time preview and editing, user login and subscription management, social sharing, and a teaching course entry interface.

[0007] Further improvements are made in the following ways: The industry template library module includes template data for clothing display, jewelry details, intangible cultural heritage narrative, food aesthetics, 3C products, beauty product trials, home life, and outdoor sports. The template data for each category has preset shot structure, background music, subtitle style, and transition logic.

[0008] Further improvements include: the one-click video creation module comprises a material intelligent analysis unit, a shot intelligent editing unit, a background music matching unit, a dynamic subtitle generation unit, and a transition effects compositing unit. The material intelligent analysis unit is used to perform subject recognition, image quality assessment, and keyframe extraction on user-uploaded images or videos. The shot intelligent editing unit is used to automatically identify the subject based on the material content and perform shot segmentation, sorting, and duration adjustment. The background music matching unit is used to automatically match and synthesize background music based on template mood tags. The dynamic subtitle generation unit is used to generate synchronized subtitles based on speech recognition or template presets. The transition effects compositing unit is used to automatically insert smooth transition animation effects between shots.

[0009] Further improvements are made in that: the intelligent material analysis unit includes a multi-target recognition subunit, a picture quality evaluation subunit, and a keyframe extraction subunit. The multi-target recognition subunit is used to simultaneously identify multiple product subjects in the picture and prioritize them. The picture quality evaluation subunit is used to score the clarity, lighting conditions, and composition rationality of the material and automatically select the best material segment. The keyframe extraction subunit is used to extract representative picture frames from the video material as shot switching nodes.

[0010] Further improvements are made in that: the virtual live streaming assistant module includes a real-time keying unit, a virtual background fusion unit, and an output unit. The real-time keying unit achieves real-time separation of the person and the background based on a deep learning model. The virtual background fusion unit is used to seamlessly fuse the keyed person with the selected high-definition virtual background. The output unit is used to directly push the synthesized image to a third-party live streaming platform or local recording.

[0011] Further improvements include: the virtual live streaming assistant module also includes a virtual background switching unit, used for switching virtual backgrounds in different scenarios, real-time background blurring and light and shadow adjustment, including product display scenarios, brand launch scenarios, and holiday promotion scenarios.

[0012] Further improvements are made in the following ways: The creator community module includes a template upload and review unit, a template trading unit, a popularity recommendation unit, and a template usage feedback unit. The template upload and review unit is used for users to upload original templates and implements template review functions. The template trading unit is used for pricing, purchasing, and revenue settlement of user-created templates. The popularity recommendation unit is used to sort and recommend templates based on download volume, rating, and usage frequency. The template usage feedback unit is used to provide template creators with download volume, user profiles, and usage heatmap analysis.

[0013] Further improvements are made in that: the training and teaching module includes a course content storage unit, a learning planning unit, a practical exercise unit, and a skills certification unit. The course content storage unit is used to store short video creation skills teaching courses. The learning planning unit is used to recommend personalized learning paths based on user level and interests. The practical exercise unit is used for users to directly call templates for practical exercises during the learning process. The skills certification unit is used to assess users who complete the course and issue platform certification certificates.

[0014] Further improvements include: the short video creation skills mentioned include, but are not limited to, camera language, editing techniques, use of light and shadow, and writing viral copy; the skill certification unit is linked to the creator community module's data, and users who obtain certification certificates will be given priority to become platform-contracted template creators, thereby receiving traffic support and unlocking advanced template usage permissions. The beneficial effects of this invention are as follows: This invention realizes full automation from material uploading to final output, significantly reducing the technical threshold and time cost of short video production. The industry templates are derived from real service cases, ensuring the high adaptability and market conversion rate of the templates. Meanwhile, the virtual live streaming assistant module uses AI real-time keying technology to enable micro and small enterprises and individual creators to obtain high-quality virtual live streaming effects without the need for professional equipment; Furthermore, the introduction of the creator community has built a user-driven, healthy content ecosystem, which not only continuously injects fresh creative resources into the platform, but also provides creators with monetization channels. This has improved internal delivery efficiency, expanded external user coverage, and formed a self-evolving and scalable technology product flywheel. Attached Figure Description

[0015] Figure 1 This is a system architecture diagram of the present invention. Detailed Implementation

[0016] To enhance understanding of the present invention, the present invention will be further described in detail below with reference to embodiments. These embodiments are only used to explain the present invention and do not constitute a limitation on the scope of protection of the present invention.

[0017] according to Figure 1 As shown, this embodiment provides an intelligent video creation system that supports UGC communities, transitions, and virtual live streaming. It includes a front-end display and interaction module, an industry template library module, a one-click video creation module, a virtual live streaming assistant module, a creator community module, and a training and teaching module integrated into a WeChat mini program. By integrating these modules into the same WeChat mini program, each module shares the template library and user data, achieving an integrated "learn-practice-use-create" experience. This solves the problems of fragmented functions, lack of data interoperability, and high learning costs associated with traditional video production tools.

[0018] The front-end display and interaction module provides template selection, material upload, video generation and preview functions, including template category browsing, intelligent search, one-click material upload, real-time preview and editing, user login and subscription management, social sharing, and a teaching course entry interface to guide users into the training and teaching module; The template category browsing interface allows users to filter by industry category, usage scenario, popularity, and latest releases; the intelligent search interface allows for keyword and image searches, and users can upload reference images to find similar templates; the one-click material upload interface allows for batch uploads and real-time upload progress display; real-time preview and editing allows for real-time preview and parameter fine-tuning during the generation process; the user login and subscription management interface allows users to log in with WeChat with one click, subscribe to memberships, make payments, and query historical records; the social sharing interface allows users to share generated videos to WeChat friends, Moments, Douyin, Kuaishou, and other platforms with one click.

[0019] The industry template library module is interconnected with the front-end display and interaction module to store industry-customized template data from real service cases, including template data for clothing display, jewelry details, intangible cultural heritage narrative, food aesthetics, 3C products, beauty trial, home life, and outdoor sports. Each category of template data has preset shot structure, background music, subtitle style and transition logic.

[0020] By building an industry template library, the templates are directly derived from real service cases (such as clothing factories, jewelry workshops, intangible cultural heritage workshops, etc.), with preset shot structures, music moods, subtitle styles and transition logic, ensuring that the templates are "ready to use out of the box" without the need for secondary adjustments by users. This solves the problem that template designers lack a deep understanding of the vertical industry of consumer goods, and that the templates are difficult to use directly in professional scenarios such as clothing texture display, jewelry detail presentation, and intangible cultural heritage narrative.

[0021] The one-click video creation module is interconnected with the front-end display and interaction module and the industry template library module. Based on the AI ​​one-click video creation engine, it automatically completes video editing, background music, letter generation and transition synthesis according to the template selected by the user and the uploaded materials. It includes a material intelligent analysis unit, a shot intelligent editing unit, a background music matching unit, a dynamic subtitle generation unit and a transition effect synthesis unit. The intelligent material analysis unit is used to perform subject recognition, image quality assessment, and keyframe extraction on user-uploaded images or videos. It includes a multi-target recognition subunit, an image quality assessment subunit, and a keyframe extraction subunit. The multi-target recognition subunit is used to simultaneously identify multiple product subjects in the image and prioritize them. The image quality assessment subunit is used to score the clarity, lighting conditions, and composition rationality of the material and automatically select the best material segment. The keyframe extraction subunit is used to extract representative image frames from the video material as shot switching nodes. The intelligent editing unit is used to automatically identify the subject based on the content of the footage and to perform shot segmentation, sorting, and duration adjustment. The background music matching unit is used to automatically match and synthesize background music based on the template's mood tags, and intelligently align the music's climax with the scene's turning point. The dynamic subtitle generation unit is used to generate synchronized subtitles based on speech recognition or template presets, while the transition effects compositing unit is used to automatically insert smooth transition animation effects between shots.

[0022] The one-click video editing module can automatically insert transition effects that match the template style between shots and supports intelligent adjustment of transition duration based on the content of the scene; it can automatically perform resolution enhancement, color correction and noise reduction on the output video; it can automatically adjust the video resolution, aspect ratio and encoding format according to the target publishing platform; it can also automatically and intelligently dub the text generated by the dynamic subtitle generation unit, and automatically generate AI voice narration based on the template style and video content, with different timbre, speech speed and language to choose from.

[0023] The one-click video creation module can also process videos in batches. Users can upload multiple product materials at once, and the system will automatically generate an independent short video for each product, and support batch export and packaged download.

[0024] This application introduces intelligent image cropping and composition optimization algorithms into the AI ​​one-click video engine, which automatically identifies the main subject in the material (such as jewelry details and clothing textures) and performs adaptive cropping for different screen formats such as Douyin, Kuaishou, and Video Accounts, ensuring that the output content conforms to the platform's recommendation mechanism. This solves the problem that traditional video engines have fixed editing rules and cannot perform intelligent cropping and composition optimization based on the main subject of the material (such as products or people).

[0025] The AI-powered one-click video creation engine utilizes a template library derived from real-world service cases (clothing, jewelry, intangible cultural heritage, food, 3C products, etc.), highly matching the visual marketing scenarios of consumer goods. Existing technologies use universal templates, designed uniformly by platform designers. The engine achieves intelligent processing through multi-unit collaboration, while traditional solutions involve splicing materials, selecting background music and subtitle styles—time-consuming and cumbersome. The AI-powered one-click video creation engine automatically identifies the main subject, enabling precise cropping and composition adjustments to ensure visual focus, something traditional tools cannot intelligently optimize. AI-powered one-click video creation... The engine can output finished videos that conform to the specifications of different platforms with one click, without the need for secondary editing. In contrast, traditional tools can only output finished videos in fixed landscape or portrait orientations, requiring secondary processing for different platforms. The AI ​​one-click video engine is simple to operate, requiring only two steps: uploading materials and selecting a template. The AI ​​automatically completes all post-production, and the output video can be published directly. In contrast, users of traditional tools still need to select materials, templates, and manually adjust duration and music. Traditional tools are isolated and can only complete video production, while the AI ​​one-click video engine can be linked with other integrated modules to form a multi-functional and intelligent video tool.

[0026] The virtual live streaming assistant module is interconnected with the front-end display and interaction module to perform real-time fusion output of people and virtual backgrounds through AI keying technology. It includes a real-time keying unit, a virtual background fusion unit, and an output unit, as well as a virtual background switching unit. The real-time keying unit uses a deep learning model to separate people from the background in real time, and can finely process complex edges such as hair strands and transparent objects. The virtual background blending unit is used to seamlessly blend the keyed-out figure with the selected high-definition virtual background, and can automatically simulate background depth of field, match light and shadow and correct color consistency. The output unit is used to directly push the synthesized image to a third-party live streaming platform or local recording, and supports multi-camera signal access and split-screen display; The virtual background switching unit is used for switching virtual backgrounds in different scenarios, real-time background blurring and lighting adjustment, including product display scenarios, brand launch scenarios, and holiday promotion scenarios.

[0027] The virtual live streaming assistant module can also add real-time beautification, filters, and special effects to the cut-out figures; it can overlay 3D models or high-definition images of products on the live stream screen and supports rotation, scaling, and angle switching operations.

[0028] The addition of the virtual live streaming assistant module provides a multi-scene virtual background library that users can switch with one click. By integrating the virtual live streaming assistant module into the WeChat mini program, it is possible to separate the person from the background in real time using only the mobile phone camera and AI keying technology, without the need for a green screen. This reduces equipment costs, and the background library is optimized for categories such as clothing, jewelry, and food to highlight the texture of the products. This integrated WeChat mini program is easy to use; users only need to select the corresponding background based on the product to start a live stream.

[0029] Current technologies rely on offline post-production for application scenarios, making real-time switching during live streaming impossible. The creator community module and the industry template library module are interconnected, allowing users to upload, share, evaluate, and trade original templates, forming a UGC content ecosystem, including a template upload and review unit, a template trading unit, a popular recommendation unit, and a template usage feedback unit; The template upload and review unit is used for users to upload original templates and for template review functions. The review content includes copyright compliance, technical compatibility, and content quality. The template trading unit is used for pricing, purchasing, and revenue settlement of user-created templates; The popularity recommendation unit is used to sort and recommend templates based on download volume, rating, and usage frequency. Specifically, it uses a collaborative filtering algorithm to sort templates and make personalized recommendations based on multi-dimensional data such as download volume, rating, usage frequency, and user feedback. The template usage feedback unit provides template creators with download data, user profiles, and usage heatmap analysis, and can generate a brief report for creators to view.

[0030] The creator community module also has an incentive function, which distributes points, cash rewards or traffic support to creators based on template usage data; it also supports interaction between creators and users.

[0031] The creator community supports users to upload and trade original templates, forming a UGC-driven self-circulating ecosystem. The template update speed is highly synchronized with user needs, solving the problem of traditional tool systems relying on the platform's R&D team for template updates and resulting in slow iteration.

[0032] The training and teaching module is connected to the front-end display and interaction module and the industry template library module to provide users with short video creation teaching courses, practical exercises and skills certification functions, including course content storage unit, learning planning unit, practical exercise unit and skills certification unit; The course content storage unit is used to store short video creation skills teaching courses, including but not limited to courses on camera language, editing techniques, use of light and shadow, and writing viral copy. The learning planning unit is used to recommend personalized learning paths based on the user's level and interests; The practical training unit is linked with the one-click video production module, allowing users to directly call up templates for practical exercises during the learning process. It also has a one-click comparison function, which simultaneously displays a split-screen comparison of the user's practical results and the standard template video, helping users to identify differences and areas for improvement. The skills certification unit is used to assess users who complete the course and issue platform certification certificates. It can be linked with the creator community module data. Users who obtain certification certificates will be given priority to become contracted template creators on the platform, in order to receive traffic support and unlock advanced template usage rights.

[0033] All the aforementioned modules' data are aggregated in the backend management system, enabling unified management and analysis of the template library, user behavior data, transaction records, and system logs.

[0034] The backend management system includes template management, user management, transaction management, data analysis, and system monitoring. Template management is used for adding, deleting, modifying, and querying templates, adjusting categories, and setting popularity weights. User management is used for user behavior tracking, profile analysis, hierarchical management, and permission settings. Transaction management is used for order processing, revenue settlement, invoice management, and refund approval. Data analysis is used for multi-dimensional statistical analysis and visualization of system operation data, user behavior data, and template usage data. System monitoring is used for real-time monitoring of system operation status, API call frequency, anomaly alarms, and log auditing.

[0035] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent video creation system supporting UGC communities, transitional video production, and virtual live streaming, characterized in that: The system includes a front-end display and interaction module integrated into WeChat Mini Programs, an industry template library module, a one-click video creation module, a virtual live streaming assistant module, a creator community module, and a training and teaching module. The front-end display and interaction module provides template selection, material upload, video generation, and preview functions. The industry template library module stores industry-customized template data from real service cases. The one-click video creation module, based on an AI one-click video creation engine, automatically completes video editing, music addition, letter generation, and transition compositing based on the user's selected template and uploaded materials. The virtual live streaming assistant module uses AI keying technology to perform real-time fusion output of characters and virtual backgrounds. The creator community module allows users to upload, share, evaluate, and trade original templates, forming a UGC content ecosystem. The training and teaching module provides users with short video creation tutorials, practical exercises, and skills certification functions.

2. The intelligent video creation system supporting UGC communities, transitional video production, and virtual live streaming as described in claim 1, characterized in that: The front-end display and interaction module includes template category browsing, intelligent search, one-click material upload, real-time preview and sharing, user login and subscription management, social sharing, and teaching course entry interface.

3. The intelligent video creation system supporting UGC communities, transitional video production, and virtual live streaming as described in claim 1, characterized in that: The industry template library module includes template data for clothing display, jewelry details, intangible cultural heritage narrative, food aesthetics, 3C products, beauty product trials, home life, and outdoor sports. Each category of template data has preset shot structure, background music, subtitle style, and transition logic.

4. The intelligent video creation system supporting UGC communities, transitional video production, and virtual live streaming as described in claim 1, characterized in that: The one-click video creation module includes a material intelligent analysis unit, a shot intelligent editing unit, a background music matching unit, a dynamic subtitle generation unit, and a transition effects compositing unit. The material intelligent analysis unit is used to perform subject recognition, image quality assessment, and keyframe extraction on user-uploaded images or videos. The shot intelligent editing unit is used to automatically identify the subject based on the material content and perform shot segmentation, sorting, and duration adjustment. The background music matching unit is used to automatically match and synthesize background music based on template mood tags. The dynamic subtitle generation unit is used to generate synchronized subtitles based on speech recognition or template presets. The transition effects compositing unit is used to automatically insert smooth transition animation effects between shots.

5. The intelligent video creation system supporting UGC communities, transitional video production, and virtual live streaming as described in claim 4, characterized in that: The intelligent material analysis unit includes a multi-target recognition subunit, a picture quality assessment subunit, and a keyframe extraction subunit. The multi-target recognition subunit is used to simultaneously identify multiple product subjects in the picture and prioritize them. The picture quality assessment subunit is used to score the clarity, lighting conditions, and composition rationality of the material and automatically select the best material segment. The keyframe extraction subunit is used to extract representative picture frames from the video material as shot switching nodes.

6. The intelligent video creation system supporting UGC communities, transitional video production, and virtual live streaming as described in claim 1, characterized in that: The virtual live streaming assistant module includes a real-time keying unit, a virtual background blending unit, and an output unit. The real-time keying unit uses a deep learning model to separate the person from the background in real time. The virtual background blending unit is used to seamlessly blend the keyed person with the selected high-definition virtual background. The output unit is used to directly push the synthesized image to a third-party live streaming platform or local recording.

7. The intelligent video creation system supporting UGC communities, transitional video production, and virtual live streaming as described in claim 1, characterized in that: The virtual live streaming assistant module also includes a virtual background switching unit, which is used for switching virtual backgrounds in different scenarios, real-time background blurring and light and shadow adjustment, including product display scenarios, brand launch scenarios, and holiday promotion scenarios.

8. The intelligent video creation system supporting UGC communities, transitional video production, and virtual live streaming according to claim 1, characterized in that: The creator community module includes a template upload and review unit, a template trading unit, a popularity recommendation unit, and a template usage feedback unit. The template upload and review unit is used for users to upload original templates and implements template review functions. The template trading unit is used for pricing, purchasing, and revenue settlement of user-created templates. The popularity recommendation unit is used to sort and recommend templates based on download volume, rating, and usage frequency. The template usage feedback unit is used to provide template creators with download volume, user profiles, and usage heatmap analysis.

9. The intelligent video creation system supporting UGC communities, transitional video production, and virtual live streaming according to claim 1, characterized in that: The training module includes a course content storage unit, a learning planning unit, a practical exercise unit, and a skills certification unit. The course content storage unit stores short video creation skills training courses. The learning planning unit recommends personalized learning paths based on user level and interests. The practical exercise unit allows users to directly call templates for practical exercises during the learning process. The skills certification unit assesses users who complete the course and issues platform certification certificates.

10. The intelligent video creation system supporting UGC communities, transitional video production, and virtual live streaming according to claim 9, characterized in that: The short video creation skills mentioned include, but are not limited to, camera language, editing techniques, use of light and shadow, and writing viral copy. The skills certification unit is linked to the creator community module, and users who obtain certification certificates will be given priority to become contracted template creators on the platform to receive traffic support and unlock advanced template usage rights.

Images