Short video editing method and system based on artificial intelligence
Through a short video editing system based on artificial intelligence, deep learning and machine learning technology are used to automatically identify key video elements and themes, and dynamically adjust video parameters, solving the problem of inefficiency of traditional manual editing and achieving efficient and personalized video production. Lower the threshold for video creation and improve user experience.
Patent Information
- Application Number
- CN202411118834.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-15
- Publication Date
- 2024-11-26
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The traditional manual video editing method is inefficient and is limited by manual subjective judgment, making it difficult to meet users' needs for efficient and personalized short videos.
Design a short video editing method and system based on artificial intelligence, including video processing unit, automatic editing unit and video output unit. Using deep learning algorithms and machine learning technology, it automatically recognizes key video elements and themes, dynamically adjusts video parameters, and provides Personalized editing recommendations and optimize content through user feedback.
It greatly reduces the difficulty of user operation, reduces the time and cost of video post-production, ensures video quality, meets user needs, provides personalized and professional video production services, and improves user experience.
Smart Images

Figure CN119031197A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the field of computer technology, and in particular relates to a short video editing method and system based on artificial intelligence. Background Art
[0002] With the rapid development of mobile Internet and social media, short video content creation has become an important way of information dissemination. However, traditional manual editing methods are inefficient and limited by human subjective judgment, making it difficult to meet users' needs for efficient and personalized short videos. Video editing software usually requires users to have certain technical knowledge and editing skills, and some existing "one-click" video editing tools often cannot meet personalized and diversified needs. Therefore, the development of an automatic short video editing method and system based on artificial intelligence is of great practical significance. Summary of the invention
[0003] The purpose of the present invention is to provide a short video editing method and system based on artificial intelligence. By setting an automatic editing unit, the problem that the existing manual editing method is inefficient and is limited by human subjective judgment, making it difficult to meet users' needs for high-efficiency and personalized short videos is solved.
[0004] In order to solve the above technical problems, the present invention is achieved through the following technical solutions:
[0005] The present invention is a short video editing method and system based on artificial intelligence, including a video processing unit, an automatic editing unit and a video output unit, wherein the video processing unit includes a video input module, a preprocessing module, a video analysis module, bullet screen surge recognition, highlight segment recognition, important scene recognition, key frame recognition and an AI preference learning module; the video input module is unidirectionally connected to the preprocessing module; the preprocessing module is unidirectionally connected to the video analysis module; the video analysis module is unidirectionally connected to key frame recognition, important scene recognition, highlight segment recognition and bullet screen surge recognition respectively; the key frame recognition, important scene recognition, highlight segment recognition and bullet screen surge recognition are unidirectionally connected to the AI preference learning module respectively, and the automatic editing unit includes an intelligent parameter module, video duration, playback speed, video quality, volume adjustment, sound quality, rough cutting, segmentation and splicing, scene transition, special effect addition, filter addition, subtitle addition, background music addition and a user interaction module; the AI preference learning module is unidirectionally connected to the intelligent parameter module; the intelligent parameter module is respectively connected to the video duration , playback speed, video quality, volume adjustment and sound quality are unidirectionally connected; the video duration, playback speed, video quality, volume adjustment and sound quality are unidirectionally connected to the rough cut respectively; the rough cut is unidirectionally connected to segmentation and splicing, transition, special effect addition, filter addition, subtitle addition and background music addition respectively; the segmentation and splicing, transition, special effect addition, filter addition, subtitle addition and background music addition are unidirectionally connected to the user interaction module respectively, and the video output unit includes a copyright detection module, a local rendering module, a multi-platform adaptation module, a cloud storage module, a video platform and a user feedback module; the user interaction module is unidirectionally connected to the copyright detection module; the local rendering module is unidirectionally connected to the multi-platform adaptation module; the multi-platform adaptation module is unidirectionally connected to the video output module; the video output module is unidirectionally connected to the cloud storage module and the video platform respectively; the video platform is unidirectionally connected to the user feedback module; the user feedback module includes likes, comments, favorites, sharing and viewing time; the user feedback module is unidirectionally connected to the AI preference learning module.
[0006] As a preferred technical solution, the video input module is responsible for receiving original video signals from different sources and parsing these signals to correctly identify and process video signals. The preprocessing module is responsible for preprocessing the parsed video material signals, denoising, de-shaking, color correction, etc., so as to improve the subsequent video viewing experience. The video analysis module is responsible for using deep learning algorithms to perform content analysis on uploaded videos, including scene recognition, object detection, action recognition, and speech-to-text conversion, etc., to extract key elements and themes of the video. The key frame recognition is responsible for automatically selecting frames that can represent important features of the video content from the video sequence. These frames usually contain the main events and important visual information in the video. The important scene recognition is responsible for using the convolutional neural network technology in deep learning to perform image analysis on each frame of the video to identify key elements in the picture such as faces, objects, Scenery, distinguish the background and foreground in the video, focus on the main activities, and capture important actions and events more accurately. The highlight segment recognition is responsible for determining and identifying highlight segments, such as the amount of activity in the picture, the volume peak in the audio, the audience reaction, etc., by analyzing the sound track to identify the exciting music climax or sudden sound changes, which often accompany the highlight segments, so as to identify the highlight segments. The barrage surge recognition is responsible for monitoring and analyzing the barrage activities in the video, especially finding the peak of the number of barrages and emotions. This module uses AI technology to help understand the audience's behavior and reaction, and provide data support for the optimization, recommendation and editing of video content. The AI preference learning module is responsible for analyzing and learning users' preferences for video content, and provides personalized video clip editing recommendations through big data and machine learning technology, while continuously optimizing the learning model to better perform clip editing analysis according to user preferences.
[0007] As a preferred technical solution, the intelligent parameter module is responsible for maintaining and optimizing various parameter settings of the video, including the basic parameters of video processing resolution, frame rate, and encoding format. According to the requirements of the video content and the target platform, the parameters are dynamically adjusted to optimize the video quality and playback effect to ensure efficient operation of the system and high-quality video output. The video duration is responsible for determining the ideal length of the video to optimize the viewing experience and content dissemination efficiency, and analyzing the viewing habits of users on the platform, including the average viewing time, preferred content types, etc., to determine the optimal video length. The playback speed is responsible for adjusting and optimizing the playback speed of the video to improve the user's viewing experience and content delivery efficiency. The video quality is responsible for ensuring the clarity and The stability meets the standards. It detects quality problems such as blur and jitter through algorithms and automatically optimizes them. It monitors the color performance in the video, including color cast, saturation and contrast, to ensure color accuracy and consistency and provide a better visual experience. The volume adjustment is responsible for ensuring that the volume of the generated and published video content is moderate and balanced, and meets the audience's auditory experience requirements. The volume of different sections in the video is uniformly adjusted to ensure that the volume of the entire video is consistent and avoid excessive volume fluctuations that affect the viewing experience. The sound quality is responsible for ensuring that the sound effects of the generated and published video content meet high standards and provide the best auditory experience to avoid blur or distortion. The rough cut is responsible for screening and organizing content from the original material and, based on the marked highlights, The editing software performs preliminary arrangement and combination on the timeline to form the basic structure of the video, ensuring that the rough cut version narrates the story in a logical order, maintaining narrative coherence and attractiveness, and laying the foundation for subsequent intelligent editing. The segmentation and splicing is responsible for accurately cutting and cleverly splicing the rough-cut video clips, removing unnecessary clips or pictures according to the story line and rhythm determined in the rough cut stage, making the story more compact and fascinating, so as to form a smooth, coherent and attractive final video. The transition is responsible for handling the smooth switching between video clips. These transition effects can smoothly connect different scenes, making the video content more coherent and natural, and enhancing the overall look and feel and narrative effect of the video. The special effects addition is responsible for adding various special effects to the video content. The video clips are divided into two types: the first type is the video clip, the second type is the video clip, and the third type is the video clip. The third type is the video clip that is edited by the user. The fourth type is the video clip that is edited by the user. The fifth type is the video clip that is edited by the user. The fifth type is the video clip that is edited by the user. The fifth type is the video clip that is edited by the user. The fifth type is the video clip that is edited by the user. The fifth type is the video clip that is edited by the user.The background music adding module is responsible for selecting music that matches the content, emotion and atmosphere of the video, and performing necessary editing of the music according to the length and structure of the video, such as cropping, fading in and out, to match the video, and adjusting the emotional changes of the music to make it consistent with the emotional fluctuations of the video content, thereby enhancing the emotional experience of the audience. The user interaction module is responsible for providing a visual interface to facilitate users to adjust editing strategies and view editing results. Users can adjust editing parameters such as editing speed, special effects type, etc. according to their own needs, thereby increasing users' editing habits.
[0008] As a preferred technical solution, the copyright detection module is responsible for ensuring that all materials used in the video (such as music, video clips, images, etc.) comply with the requirements of copyright laws, comparing them with the copyright database, checking the copyright status of these materials, and determining whether copyright permission is required or whether they belong to the public domain. When a possible copyright infringement is found, the video producer or editor is notified immediately, and relevant copyright information and solution suggestions are provided. The local rendering module is responsible for converting the edited video project into the final video file through the rendering process, and uses synthesis, special effects application, encoding and other steps to output high-quality videos that meet playback standards. Detailed rendering setting options are provided, including resolution, bit rate, frame rate and encoding format, etc., so that users can choose the most appropriate rendering quality and file size according to their needs. The multi-platform adaptation module is responsible for ensuring that the video content can be correctly displayed and played on various devices and platforms, maintaining the aspect ratio of the video content to adapt to the display characteristics of different platforms, and avoiding stretching or deformation. , ensuring the consistency of visual effects, including different operating systems, browsers and social media. The video output module is responsible for outputting the edited and rendered video files in the correct format and quality. According to the target audience's equipment and network conditions, the appropriate resolution is selected, such as 4K, 1080p, 720p, etc., and the bit rate is adjusted to control the video quality and file size to ensure smooth playback without losing quality, especially when considering the network playback environment, to ensure that the video can achieve the best viewing effect in different usage scenarios. The video platform is responsible for the main carrier and distribution channel for the edited videos, so that more viewers can watch the videos made by users. The cloud storage module is responsible for storing the edited videos for subsequent viewing. The user feedback module is responsible for collecting user feedback on the generated short videos, including likes, favorites, comments, shares and viewing time, so as to continuously collect users' video viewing preferences, so as to train the AI preference learning module in the future.
[0009] The present invention has the following beneficial effects:
[0010] By setting up an automatic editing unit, the present invention greatly reduces the difficulty of user operation compared to traditional editing software. Users can produce high-quality video content without professional video editing skills. Through the automated editing process, the time and cost of video post-production are greatly reduced, while ensuring that the video content has high professional quality to meet user needs. Users can easily create high-quality video content without professional video editing skills, which reduces the threshold for video creation and enables more users to participate in video creation.
[0011] The present invention sets up a user interaction module, which allows users to preview the editing effect in real time by providing an interactive editing support module to users, and can adjust it according to the user's own editing habits. This dynamic editing process enhances the user's sense of participation and creativity, optimizes artificial intelligence editing preferences, and thus provides users with more personalized and professional video production services, thereby enhancing the user's experience;
[0012] The present invention sets up a user feedback module, which transmits the data of likes, comments, favorites, shares and viewing time collected from the published videos to the AI preference learning module. These data can not only help the artificial intelligence better understand and meet user needs and improve the quality and accuracy of content recommendations, but also promote the optimization of edited content, so that the AI preference learning module can predict the video types or themes that may be popular in the future based on the current interactive data trends, thereby intelligently editing trend content and preferred themes.
[0013] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS
[0014] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for describing the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.
[0015] Figure 1 The present invention provides an artificial intelligence-based short video editing method and system. DETAILED DESCRIPTION
[0016] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0017] See also Figure 1 As shown, the present invention is a short video editing method and system based on artificial intelligence, including a video processing unit, an automatic editing unit and a video output unit;
[0018] The video processing unit includes a video input module, a preprocessing module, a video analysis module, bullet screen surge recognition, highlight clip recognition, important scene recognition, key frame recognition and an AI preference learning module; the video input module is unidirectionally connected to the preprocessing module; the preprocessing module is unidirectionally connected to the video analysis module; the video analysis module is unidirectionally connected to key frame recognition, important scene recognition, highlight clip recognition and bullet screen surge recognition; key frame recognition, important scene recognition, highlight clip recognition and bullet screen surge recognition are unidirectionally connected to the AI preference learning module respectively, and the automatic editing unit includes an intelligent parameter module, video duration, playback speed, video quality, volume adjustment, sound quality, rough cutting, segmentation and splicing, transition, special effects addition, filter addition, subtitle addition, background music addition and user interaction module; the AI preference learning module is unidirectionally connected to the intelligent parameter module; the intelligent parameter module is unidirectionally connected to video duration, playback speed, video quality, volume adjustment and sound quality respectively One-way connection; video length, playback speed, video quality, volume adjustment and sound quality are one-way connected to the rough cut respectively; the rough cut is one-way connected to segmentation and splicing, transition, special effect addition, filter addition, subtitle addition and background music addition respectively; segmentation and splicing, transition, special effect addition, filter addition, subtitle addition and background music addition are one-way connected to the user interaction module respectively, and the video output unit includes a copyright detection module, a local rendering module, a multi-platform adaptation module, a cloud storage module, a video platform and a user feedback module; the user interaction module is one-way connected to the copyright detection module; the local rendering module is one-way connected to the multi-platform adaptation module; the multi-platform adaptation module is one-way connected to the video output module; the video output module is one-way connected to the cloud storage module and the video platform respectively; the video platform is one-way connected to the user feedback module; the user feedback module includes likes, comments, favorites, sharing and viewing time; the user feedback module is one-way connected to the AI preference learning module.
[0019] The video input module is responsible for receiving raw video signals from different sources and parsing these signals in order to correctly identify and process the video signals. The preprocessing module is responsible for preprocessing the parsed video material signals, denoising, de-shaking, color correction, etc., so as to improve the subsequent video viewing experience. The video analysis module is responsible for using deep learning algorithms to perform content analysis on the uploaded videos, including scene recognition, object detection, action recognition, and speech-to-text, etc., to extract the key elements and themes of the video. Key frame recognition is responsible for automatically selecting frames from the video sequence that can represent the important features of the video content. These frames usually contain the main events and important visual information in the video. Important scene recognition is responsible for using the convolutional neural network technology in deep learning to perform image analysis on each frame of the video, identify key elements in the picture such as faces, objects, and scenery, and distinguish between the key elements in the video. The background and foreground of the video can be distinguished, and the main activities can be focused on to capture important actions and events more accurately. The highlight segment recognition is responsible for determining the highlight segments, such as the amount of activity in the picture, the volume peak in the audio, the audience reaction, etc. It analyzes the sound track to identify the exciting music climax or sudden sound changes, which often accompany the highlight segments, so as to identify the highlight segments. The barrage surge recognition is responsible for monitoring and analyzing the barrage activities in the video, especially finding the peak of the number of barrages and emotions. This module uses AI technology to help understand the audience's behavior and reaction, and provide data support for the optimization, recommendation and editing of video content. The AI preference learning module is responsible for analyzing and learning users' preferences for video content, and provides personalized video clip editing recommendations through big data and machine learning technology, while continuously optimizing the learning model to better perform clip editing analysis based on user preferences.
[0020] The intelligent parameter module is responsible for maintaining and optimizing various parameter settings of the video, including the basic parameters of video processing resolution, frame rate, and encoding format. According to the requirements of the video content and the target platform, the parameters are dynamically adjusted to optimize the video quality and playback effect to ensure the efficient operation of the system and high-quality video output. The video duration is responsible for determining the ideal length of the video to optimize the viewing experience and content dissemination efficiency. It analyzes the viewing habits of users on the platform, including the average viewing time, preferred content types, etc., to determine the optimal video length. The playback speed is responsible for adjusting and optimizing the playback speed of the video to improve the user's viewing experience and content delivery efficiency. The video quality is responsible for ensuring that the clarity and stability of the video content meet the standards, and detects blur, jitter and other quality issues through algorithms. Problems are detected and automatically optimized. The color performance in the video is monitored, including color cast, saturation and contrast, to ensure color accuracy and consistency and provide a better visual experience. Volume adjustment is responsible for ensuring that the volume of generated and published video content is moderate and balanced, and meets the audience's auditory experience requirements. The volume of different sections in the video is uniformly adjusted to ensure the consistency of the volume of the entire video and avoid excessive volume fluctuations that affect the viewing experience. Sound quality is responsible for ensuring that the sound effects of generated and published video content meet high standards and provide the best auditory experience to avoid blur or distortion. Rough cut is responsible for screening and organizing content from raw materials, and performing preliminary arrangement and combination on the timeline of the editing software based on the marked highlights to form the basic structure of the video to ensure The rough cut version narrates the story in a logical order, maintaining narrative coherence and appeal, laying the foundation for subsequent intelligent editing. Segmentation and splicing is responsible for accurately cutting and cleverly splicing the rough-cut video clips. According to the story line and rhythm determined in the rough cut stage, unnecessary clips or pictures are removed to make the story more compact and fascinating, so as to form a smooth, coherent and attractive final video. Transition is responsible for handling the smooth switching between video clips. These transition effects can smoothly connect different scenes, making the video content more coherent and natural, and enhancing the overall look and feel and narrative effect of the video. Special effects addition is responsible for adding various visual and audio effects to the video content. These special effects can improve the visual quality and style of the video, increase the emotion of the story, and enhance The attractiveness of the video and the viewing experience of the audience, filter addition is responsible for adding various filter effects to the video material, selecting appropriate filters according to the theme and style of the video, adjusting the brightness and lighting effects of the video, and improving the problem of insufficient light or overexposure caused by limited shooting conditions to improve the visual effect, enhance the story atmosphere, and increase the layering and depth of the picture. Subtitles are responsible for adding text information to the video content, including dialogues, commentary text, titles, tail words, etc., to enhance the information transmission and viewing experience of the video. According to the content and style of the video, create subtitles that match the theme to ensure that the text is concise and clear and can accurately convey the video information. Background music is responsible for selecting music that matches the content, emotion and atmosphere of the video, and according to the length and structure of the video,Perform necessary editing on the music, such as cropping, fading in and out, to match it with the video, adjust the emotional changes of the music to make it consistent with the emotional fluctuations of the video content, and enhance the audience's emotional experience. The user interaction module is responsible for providing a visual interface to facilitate users to adjust editing strategies and view editing results. Users can adjust editing parameters such as editing speed, special effects type, etc. according to their needs to increase users' editing habits.
[0021] The copyright detection module is responsible for ensuring that all materials used in the video (such as music, video clips, images, etc.) comply with the requirements of copyright laws, comparing them with the copyright database, checking the copyright status of these materials, and determining whether copyright permission is required or whether they belong to the public domain. When possible copyright infringement is found, the video producer or editor is notified immediately, and relevant copyright information and solution suggestions are provided. The local rendering module is responsible for converting the edited video project into the final video file through the rendering process, using synthesis, special effects application, encoding and other steps to output high-quality videos that meet playback standards. It provides detailed rendering setting options, including resolution, bit rate, frame rate and encoding format, allowing users to choose the most appropriate rendering quality and file size according to their needs. The multi-platform adaptation module is responsible for ensuring that the video content can be displayed and played correctly on various devices and platforms, maintaining the aspect ratio of the video content to adapt to the display characteristics of different platforms, avoiding stretching or deformation, and ensuring visual The consistency of the effect, including different operating systems, browsers and social media. The video output module is responsible for outputting the edited and rendered video files in the correct format and quality. According to the target audience's equipment and network conditions, select the appropriate resolution, such as 4K, 1080p, 720p, etc., adjust the bit rate to control the video quality and file size, and ensure smooth playback without losing quality, especially when considering the network playback environment, to ensure that the video can achieve the best viewing effect in different usage scenarios. The video platform is responsible for the main carriers and distribution channels for the edited videos, so that more viewers can watch the videos made by users. The cloud storage module is responsible for storing the edited videos for subsequent viewing. The user feedback module is responsible for collecting user feedback on the generated short videos, including likes, favorites, comments, shares and viewing time, so as to continuously collect users' video viewing preferences, and then train the AI preference learning module.
[0022] A specific application of this embodiment is as follows: by inputting a video into a video input module, the preprocessing module performs preliminary preprocessing such as denoising, de-shaking, and color correction on the video material; the video analysis module performs content analysis on the uploaded video, including scene recognition, object detection, action recognition, and speech-to-text conversion, etc., to extract key elements and themes of the video; after passing through several recognition modules, the highlight clips, scenes, and key frames in the video are marked and captured; the AI preference learning module analyzes and learns the user's preference for video content; and provides personalized video clip editing recommendations through big data and machine learning technology; the intelligent parameter module optimizes various parameter settings of the video; after adjusting the video duration, playback speed, video quality, volume adjustment, and sound quality, the content in the original material is screened and organized; and according to the marked highlight clips, the video is edited in the editing software. A preliminary arrangement and combination is performed on the timeline to form the basic structure of the video. The rough-cut video is then fine-cut by splitting and splicing, transitioning, adding special effects, filters, subtitles, and background music. The fine-cut video is displayed in a visual interface, which is convenient for users to adjust editing strategies and view editing results. Users can adjust editing preferences according to their needs. Users confirm the final version of the video editing and perform copyright detection. After it is found that it meets the requirements of copyright law, the edited video project is converted into the final video file through the rendering process to ensure that the video content can be correctly displayed and played on various devices and platforms. After the video platform is issued, the user feedback module is used to collect user feedback on the generated short video. Through the AI preference learning module, the quality and accuracy of content recommendations are improved to promote the optimization of editing content.
[0023] It is worth noting that in the above system embodiment, the various units included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.
[0024] In addition, those skilled in the art can understand that all or part of the steps in the above-mentioned embodiments can be completed by instructing related hardware through a program, and the corresponding program can be stored in a computer-readable storage medium.
[0025] The preferred embodiments of the present invention disclosed above are only used to help illustrate the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to the specific implementation methods described. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.
Claims
1. A short video editing method and system based on artificial intelligence, comprising a video processing unit, an automatic editing unit and a video output unit, characterized in that: The video processing unit includes a video input module, a preprocessing module, a video analysis module, bullet screen surge recognition, a highlight clip recognition, an important scene recognition, a key frame recognition and an AI preference learning module; the video input module is unidirectionally connected to the preprocessing module; the preprocessing module is unidirectionally connected to the video analysis module; the video analysis module is unidirectionally connected to the key frame recognition, the important scene recognition, the highlight clip recognition and the bullet screen surge recognition respectively; the key frame recognition, the important scene recognition, the highlight clip recognition and the bullet screen surge recognition are unidirectionally connected to the AI preference learning module respectively; The automatic editing unit includes an intelligent parameter module, video duration, playback speed, video quality, volume adjustment, sound quality, rough cut, segmentation and splicing, transition, special effect addition, filter addition, subtitle addition, background music addition and user interaction module; the AI preference learning module is unidirectionally connected to the intelligent parameter module; the intelligent parameter module is unidirectionally connected to the video duration, playback speed, video quality, volume adjustment and sound quality respectively; the video duration, playback speed, video quality, volume adjustment and sound quality are unidirectionally connected to the rough cut respectively; the rough cut is unidirectionally connected to segmentation and splicing, transition, special effect addition, filter addition, subtitle addition and background music addition respectively; the segmentation and splicing, transition, special effect addition, filter addition, subtitle addition and background music addition are unidirectionally connected to the user interaction module respectively; The video output unit includes a copyright detection module, a local rendering module, a multi-platform adaptation module, a cloud storage module, a video platform and a user feedback module; the user interaction module is unidirectionally connected to the copyright detection module; the local rendering module is unidirectionally connected to the multi-platform adaptation module; the multi-platform adaptation module is unidirectionally connected to the video output module; the video output module is unidirectionally connected to the cloud storage module and the video platform respectively; the video platform is unidirectionally connected to the user feedback module; the user feedback module includes likes, comments, favorites, shares and viewing time; the user feedback module is unidirectionally connected to the AI preference learning module.
2. The short video editing method and system based on artificial intelligence according to claim 1, characterized in that: The video input module is responsible for receiving original video signals from different sources and analyzing these signals so as to correctly identify and process the video signals; The preprocessing module is responsible for preprocessing the parsed video material signal, removing noise, de-shaking, color correction, etc., so as to improve the subsequent video viewing experience; The video analysis module is responsible for performing content analysis on the uploaded video using deep learning algorithms, including scene recognition, object detection, action recognition, and speech-to-text conversion, to extract key elements and themes of the video; The key frame recognition is responsible for automatically selecting frames that can represent important features of the video content from the video sequence. These frames usually contain the main events and important visual information in the video. The important scene recognition is responsible for using the convolutional neural network technology in deep learning to perform image analysis on each frame of the video, identify key elements in the picture such as faces, objects, and scenery, distinguish the background and foreground in the video, focus on the main activities, and more accurately capture important actions and events; The highlight segment identification is responsible for determining and identifying the highlight segments, such as the amount of activity in the picture, the volume peak in the audio, the audience reaction, etc., by analyzing the sound track to identify the exciting music climax or sudden sound changes, which often accompany the highlight segments, so as to identify the highlight segments; The barrage surge recognition module is responsible for monitoring and analyzing barrage activities in videos, especially finding the peaks of barrage quantity and emotions. This module uses AI technology to help understand the audience's behavior and reactions, and provide data support for the optimization, recommendation and editing of video content. The AI preference learning module is responsible for analyzing and learning users' preferences for video content, providing personalized video clip editing recommendations through big data and machine learning technology, and continuously optimizing the learning model to better perform clip editing analysis based on users' preferences.
3. The short video editing method and system based on artificial intelligence according to claim 2, characterized in that: The intelligent parameter module is responsible for maintaining and optimizing various video parameter settings, including basic parameters such as resolution, frame rate, and encoding format of video processing. According to the requirements of the video content and the target platform, the parameters are dynamically adjusted to optimize the video quality and playback effect, so as to ensure efficient operation of the system and high-quality video output; The video duration is responsible for determining the ideal length of the video to optimize the viewing experience and content dissemination efficiency, and analyzing the viewing habits of users on the platform, including average viewing time, preferred content types, etc., to determine the optimal video length; The playback speed is responsible for adjusting and optimizing the playback speed of the video to improve the user's viewing experience and the content delivery efficiency; The video quality is responsible for ensuring that the clarity and stability of the video content meet the standards, detecting quality problems such as blur and jitter through algorithms, and automatically optimizing processing, monitoring the color performance in the video, including color cast, saturation and contrast, to ensure color accuracy and consistency, and provide a better visual experience; The volume adjustment is responsible for ensuring that the volume of generated and published video content is moderate and balanced, and meets the audience's auditory experience requirements, and uniformly adjusts the volume of different sections in the video to ensure that the volume of the entire video is consistent, avoiding excessive volume fluctuations that affect the viewing experience; The audio quality is responsible for ensuring that the audio of the generated and published video content meets high standards and provides the best auditory experience without blur or distortion; The rough cut is responsible for selecting and organizing the content from the original material, and performing preliminary arrangement and combination on the timeline of the editing software according to the marked essential clips to form the basic structure of the video, ensuring that the rough cut version narrates the story in a logical order, maintains the narrative coherence and appeal, and lays the foundation for subsequent intelligent editing; The segmentation and splicing is responsible for accurately cutting and skillfully splicing the rough-cut video clips, removing unnecessary clips or pictures according to the story line and rhythm determined in the rough-cut stage, making the story more compact and fascinating, so as to form a smooth, coherent and attractive final video; The transition is responsible for processing the smooth switching between video clips. These transition effects can smoothly connect different scenes, making the video content more coherent and natural, and improving the overall look and feel of the video and the narrative effect; The special effects adding is responsible for adding various visual and audio special effects to the video content, which can improve the visual quality and style of the video, increase the emotions of the story, so as to enhance the appeal of the video and the viewing experience of the audience; The filter adding function is responsible for adding various filter effects to the video material, selecting appropriate filters according to the theme and style of the video, adjusting the brightness and light effects of the video, and improving the problem of insufficient light or overexposure caused by shooting conditions, so as to improve the visual effect, enhance the story atmosphere, and increase the layering and depth of the picture; The subtitles are responsible for adding text information to the video content, including dialogue, commentary, titles, tail words, etc., to enhance the information transmission and viewing experience of the video. According to the content and style of the video, subtitles that match the theme are created to ensure that the text is concise and clear and can accurately convey the video information. The background music adding is responsible for selecting music that matches the content, emotion and atmosphere of the video, performing necessary editing of the music according to the length and structure of the video, such as cropping, fading in and out, etc., to match the video, adjusting the emotional changes of the music to make it consistent with the emotional fluctuations of the video content, and enhancing the emotional experience of the audience; The user interaction module is responsible for providing a visual interface to facilitate users to adjust editing strategies and view editing results. Users can adjust editing parameters such as editing speed, special effects type, etc. according to their needs to improve users' editing habits.
4. The short video editing method and system based on artificial intelligence according to claim 3, characterized in that: The copyright detection module is responsible for ensuring that all materials used in the video (such as music, video clips, images, etc.) comply with the requirements of copyright laws, comparing them with the copyright database, checking the copyright status of these materials, determining whether copyright permission is required or whether they belong to the public domain, and immediately notifying the video producer or editor when possible copyright infringement is found, and providing relevant copyright information and solution suggestions; The local rendering module is responsible for converting the edited video project into a final video file through the rendering process, and uses synthesis, special effect application, encoding and other steps to output high-quality videos that meet playback standards. It provides detailed rendering setting options, including resolution, bit rate, frame rate and encoding format, so that users can choose the most appropriate rendering quality and file size according to their needs; The multi-platform adaptation module is responsible for ensuring that the video content can be correctly displayed and played on various devices and platforms, maintaining the aspect ratio of the video content to adapt to the display characteristics of different platforms, avoiding stretching or deformation, and ensuring consistency of visual effects, including different operating systems, browsers and social media; The video output module is responsible for outputting the edited and rendered video files in the correct format and quality, selecting the appropriate resolution, such as 4K, 1080p, 720p, etc., according to the target audience's equipment and network conditions, adjusting the bit rate to control the video quality and file size, ensuring smooth playback without losing quality, especially when considering the network playback environment, ensuring that the video can achieve the best viewing effect in different usage scenarios; The video platform is responsible for the main carrier and distribution channel of the edited videos, so that more viewers can watch the videos made by users; The cloud storage module is responsible for storing the edited video for subsequent viewing; The user feedback module is responsible for collecting user feedback on the generated short videos, including likes, favorites, comments, shares and viewing time, so as to continuously collect users' video viewing preferences and subsequently train the AI preference learning module.
Citation Information
Patent Citations
Video recommendation method and display equipment
CN114117126A
Video post-editing and video synthesis optimization method
CN116847123A
Short video production method and system based on artificial intelligence
CN117714797A
Video synthesis method and system based on semantic analysis and storage medium
CN118200665A
Social media asset portal
US20180367826A1
Cited By
Multi-angle video automatic editing method based on content defect detection
CN119383400A
Automatic editing method of multi-angle video based on content defect detection
CN119383400B
Video synthesis method and device
CN119450146A
Short video clip synthesis method based on fuzzy logic
CN119854573A
Video editing method and system based on AI intelligence
CN120186414A
