Digital media content intelligent editing and special effect generating system based on artificial intelligence

Abstract

The invention relates to the technical field of editing and special effect generation, in particular to a digital media content intelligent editing and special effect generation system based on artificial intelligence. Comprising a content understanding analysis module, an editing decision module, a special effect generation preview module, an audio processing module, a dynamic transition special effect module, a subtitle generation editing module, a virtual interaction module and a content distribution optimization module. The content understanding analysis module is used for video content identification and sentiment analysis; and the editing decision module is used for recommending an editing point and providing a dynamic editing strategy. The video content can be automatically analyzed, the editing point and the special effect adding position can be identified, the editing and special effect generation efficiency is greatly improved, the editing strategy and the special effect type can be dynamically adjusted according to the video content and the sentiment analysis result, the editing and the special effect can better meet the user requirement and the video content, and the user experience is improved. Abundant editing and special effect options are provided, and the editing and special effect generation process is simplified.

Description

Technical Field

[0001] The present invention relates to the technical field of video editing and special effect generation, and specifically to an intelligent video editing and special effect generation system for digital media content based on artificial intelligence. Background Art

[0002] Digital media refers to information carriers stored, processed, and transmitted in binary form, including various forms of media content such as text, images, audio, and video. With the popularization of the Internet and the development of technology, digital media has become one of the main ways for people to obtain information, entertainment, and communication. Content intelligent editing refers to the use of artificial intelligence technology to perform automated or semi-automated editing on digital media content. Among them, generating special effects during the editing process is also an essential step. Special effect generation refers to adding various visual or auditory effects to digital media content to enhance its expressiveness and attractiveness. Traditional special effect generation methods usually require professional special effect production software and skills, which have a relatively high threshold for ordinary users.

[0003] At the same time, generally, traditional editing methods often require manual frame-by-frame viewing of videos, manually selecting editing points, and adjusting the editing order and duration, which is both time-consuming and laborious. Content intelligent editing technology can analyze video content through algorithms, automatically identify elements such as shot transitions, character movements, and scene changes, and thus automatically generate edited versions that meet requirements according to preset editing rules or user preferences. Moreover, traditional editing methods highly rely on the experience and aesthetics of editors. Different editors may produce completely different editing effects. For long videos or a large amount of video content, manual editing takes a lot of time, is difficult to meet the demand for rapid processing, and is also difficult to dynamically adjust according to the real-time changes of video content. The editing strategy is relatively fixed.

[0004] In summary, it is necessary to propose an intelligent video editing and special effect generation system for digital media content based on artificial intelligence to solve the above problems. Summary of the Invention

[0005] The purpose of the present invention is to provide an intelligent video editing and special effect generation system for digital media content based on artificial intelligence to solve the problems raised in the above background art.

[0006] To achieve the above purpose, the present invention provides the following technical solutions:

[0007] An intelligent video editing and special effect generation system for digital media content based on artificial intelligence includes a content understanding and analysis module, an editing decision module, a special effect generation and preview module, an audio processing module, a dynamic transition special effect module, a subtitle generation and editing module, a virtual interaction module, and a content distribution and optimization module;

[0008] The content understanding and analysis module is used for video content recognition and emotion analysis;

[0009] The clip decision module is used for clip point recommendation and providing dynamic clip strategies;

[0010] The special effect generation preview module is used for real-time style transfer and special effect preview interaction;

[0011] The audio processing module is used for audio separation and sound effect addition;

[0012] The dynamic transition special effect module is used for transition recognition and transition special effect generation;

[0013] The subtitle generation and editing module is used for automatic subtitle generation and subtitle style editing;

[0014] The virtual interaction module is used for virtual character generation and interactive element addition;

[0015] The content distribution optimization module is used for platform adaptation and content optimization suggestions.

[0016] Preferably, the content understanding and analysis module further includes a video content recognition unit and an emotion analysis unit;

[0017] The video content recognition unit performs video frame content recognition through a convolutional neural network, and is used to automatically classify and label scene, character, and object elements in the video; analyzes the speech content in the video using natural language processing technology, and extracts keywords and themes;

[0018] The emotion analysis unit performs emotion tendency analysis on the speech and pictures in the video based on an LSTM-based emotion classification model, and is used to judge the emotional color of the video content; combines facial expression recognition technology to further refine the emotional expression of characters in the video.

[0019] Preferably, the clip decision module further includes a clip point recommendation unit and a dynamic clip strategy unit;

[0020] The clip point recommendation unit uses inter-frame difference analysis to identify shot transition points in the video and provides preliminary suggestions for editing; combines quantity and rhythm changes to optimize clip point recommendations to make the editing smoother and more natural;

[0021] The dynamic clip strategy unit dynamically adjusts clip strategies according to video content and emotion analysis results, including using fast editing for fast-paced videos and slow editing for slow-paced videos; realizes adaptive clip length and automatically adjusts the duration of segments according to the importance of video content.

[0022] Preferably, the special effect generation preview module further includes a real-time style transfer unit and a special effect preview interaction unit;

[0023] The real-time style transfer unit uses a style transfer model to achieve real-time conversion of video styles, including converting ordinary videos into oil painting styles and anime styles;

[0024] This unit supports users to customize styles by uploading style images for style transfer;

[0025] The special effect preview interaction unit uses augmented reality technology to achieve real-time preview of special effects. Users can directly see the special effect effects during the editing process; an interactive special effect adjustment interface is provided, and users can adjust special effect parameters through drag and zoom operations.

[0026] Preferably, the audio processing module further includes an audio separation unit and a sound effect addition unit;

[0027] The audio separation unit applies independent component analysis or U-Net network to achieve separation of audio signals, including separating background music and vocals;

[0028] This unit supports separation and processing of multi-channel audio, improving the flexibility of audio editing;

[0029] The sound effect addition unit automatically recommends and adds suitable sound effects or background music according to the video content and emotional analysis results; uses waveform superposition audio synthesis to generate custom sound effects to meet personalized needs.

[0030] Preferably, the dynamic transition special effect module further includes a transition recognition unit and a transition special effect generation unit;

[0031] The transition recognition unit analyzes the video content through a sequence-to-sequence model to automatically identify positions suitable for adding transition special effects; dynamically adjusts the type and duration of transition special effects according to the video rhythm and emotional changes;

[0032] The transition special effect generation unit applies computer graphics technology to generate rich transition special effects, including dissolve, push-pull, and rotation;

[0033] This unit supports users to customize transition special effects, providing a variety of special effect templates and parameter adjustment options.

[0034] Preferably, the subtitle generation and editing module further includes an automatic subtitle generation unit and a subtitle style editing unit;

[0035] The automatic subtitle generation unit uses the ASR speech recognition method to convert the speech content in the video into text and automatically generates subtitles;

[0036] This unit supports generation of multi-language subtitles to meet internationalization needs;

[0037] The subtitle style editing unit provides a variety of subtitle style templates, and users can select appropriate subtitle styles according to the video style;

[0038] Through this unit, it supports the custom editing of subtitle position, font, and color attributes, improving the readability and aesthetics of subtitles.

[0039] Preferably, the virtual interaction module further includes a virtual character generation unit and an interaction element adding unit;

[0040] The virtual character generation unit applies Maya modeling animation technology to create virtual characters, including animated characters and virtual anchors;

[0041] Through this unit, it supports character motion capture and expression mapping, realizing the interaction of virtual characters in a real environment;

[0042] The interaction element adding unit provides a rich library of interaction elements, including buttons, sliders, and pop-up windows. Users can add them to the video to enhance interactivity; use programming interfaces to implement the logical control and event response of interaction elements.

[0043] Preferably, the content distribution optimization module further includes a platform adaptation unit and a content optimization suggestion unit;

[0044] The platform adaptation unit automatically adjusts video resolution, frame rate, and encoding format parameters according to the specifications and requirements of different video platforms; provides a one-key publishing function, supporting simultaneous publishing to multiple video platforms;

[0045] The content optimization suggestion unit uses hot topic mining to provide optimization suggestions for video content; adjusts the content strategy in real time according to the performance data of video play volume, like count, and comment count, improving video exposure rate and user participation.

[0046] Based on the above system, the present invention also proposes an intelligent digital media content intelligent editing and special effect generation method based on artificial intelligence, including the following steps:

[0047] S1. Video content recognition and sentiment analysis

[0048] The system first receives the input video file, performs content recognition on video frames through a convolutional neural network, and automatically classifies and labels scene, character, and object elements in the video;

[0049] At the same time, uses natural language processing technology to analyze the speech content in the video, extracting keywords and themes;

[0050] Then, based on the LSTM-based sentiment classification model, performs sentiment tendency analysis on the speech and pictures in the video, judges the emotional color of the video content, and combines facial expression recognition technology to refine the emotional expression of characters;

[0051] S2. Clip Point Recommendation and Dynamic Editing Strategy Formulation

[0052] The system uses inter-frame difference analysis to identify shot transition points in the video, provides preliminary suggestions for editing, and optimizes the clip point recommendations by combining volume and rhythm changes;

[0053] According to the video content and sentiment analysis results, the reinforcement learning algorithm is applied to dynamically adjust the editing strategy, including fast editing for fast-paced videos and slow editing for slow-paced videos, to achieve adaptive editing length;

[0054] S3. Real-time Style Transfer and Special Effect Preview Interaction

[0055] The system uses a style transfer model to achieve real-time conversion of the video style. Users can choose to convert the video into an oil painting style, anime style, or upload a custom style image for transfer;

[0056] Utilize augmented reality technology to achieve real-time preview of special effects. Users can directly see the special effect effects during the editing process and adjust the special effect parameters through dragging and scaling operations;

[0057] S4. Audio Separation and Sound Effect Addition

[0058] The system applies independent component analysis or U-Net network to achieve separation of audio signals, including separating background music from human voices, and supports separation and processing of multi-channel audio;

[0059] According to the video content and sentiment analysis results, the system automatically recommends and adds suitable sound effects or background music. Users can also use waveform superposition audio synthesis to generate custom sound effects;

[0060] S5. Transition Recognition and Transition Special Effect Generation

[0061] The system analyzes the video content through a sequence-to-sequence model, automatically identifies the positions suitable for adding transition special effects, and dynamically adjusts the type and duration of the transition special effects according to the video rhythm and emotional changes;

[0062] Apply computer graphics technology to generate rich transition special effects, including dissolve, push-pull, rotation. Users can choose preset templates or customize special effect parameters;

[0063] S6. Automatic Subtitle Generation and Subtitle Style Editing

[0064] The system uses the ASR speech recognition method to convert the speech content in the video into text, automatically generates subtitles, and supports multi-language subtitle generation;

[0065] Provide a variety of subtitle style templates for users to choose from. Users can adjust subtitle positions, fonts, and color attributes according to the video style;

[0066] S7. Virtual Character Generation and Interactive Element Addition

[0067] The system applies modeling animation technology to create virtual characters, including animated figures and virtual anchors, supports character motion capture and expression mapping, and realizes the interaction between virtual characters and the real environment;

[0068] Provide a rich library of interactive elements, including buttons, sliders, and pop - up windows. Users can add them to the video to enhance interactivity and achieve logical control and event response through programming interfaces;

[0069] S8. Platform Adaptation and Content Optimization Suggestions

[0070] The system automatically adjusts video resolution, frame rate, and encoding format parameters according to the specifications and requirements of different video platforms, provides a one - click publishing function, and supports publishing to multiple video platforms simultaneously;

[0071] Use popular topic mining technology to provide optimization suggestions for video content, and adjust the content strategy in real - time according to performance data such as video play volume, like count, and comment count to improve video exposure and user engagement.

[0072] Compared with the prior art, the beneficial effects of the present invention are as follows: The present invention can automatically analyze video content, identify editing points and special effect addition positions, greatly improving the efficiency of editing and special effect generation. It can dynamically adjust editing strategies and special effect types according to video content and sentiment analysis results, making the editing and special effects more in line with user needs and video content. It also provides a rich variety of editing and special effect options, allowing users to customize according to their own preferences and needs, simplifying the process of editing and special effect generation, enabling ordinary users to easily get started, reducing the technical threshold, supporting real - time preview and adjustment of special effects. Users can directly see the effects during the editing process and perform operations such as dragging and scaling to adjust special effect parameters. Brief Description of the Drawings

[0073] Figure 1 Shows the topology diagram of the intelligent digital media content intelligent editing and special effect generation system based on artificial intelligence of the present invention;

[0074] Figure 2 Shows the flowchart of the intelligent digital media content intelligent editing and special effect generation method based on artificial intelligence of the present invention. Detailed Embodiments

[0075] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the scope of protection of the present invention.

[0076] Embodiment 1

[0077] Please refer to Figure 1 , the present invention proposes an intelligent digital media content editing and special effect generation system based on artificial intelligence, including a content understanding and analysis module, a editing decision module, a special effect generation and preview module, an audio processing module, a dynamic transition special effect module, a subtitle generation and editing module, a virtual interaction module, and a content distribution optimization module;

[0078] Among them, it should be noted that the content understanding and analysis module of this system is used for video content recognition and emotion analysis, the editing decision module of this system is used for editing point recommendation and providing dynamic editing strategies, the special effect generation and preview module of this system is used for real-time style transfer and special effect preview interaction, the audio processing module of this system is used for audio separation and sound effect addition, the dynamic transition special effect module of this system is used for transition recognition and transition special effect generation, the subtitle generation and editing module of this system is used for automatic subtitle generation and subtitle style editing, the virtual interaction module of this system is used for virtual character generation and interaction element addition, and the content distribution optimization module of this system is used for platform adaptation and content optimization suggestions.

[0079] In this embodiment, it should also be noted that the content understanding and analysis module further includes a video content recognition unit and an emotion analysis unit;

[0080] Furthermore, the video content recognition unit performs video frame content recognition through a convolutional neural network, which is used to automatically classify and label the scene, characters, and object elements in the video; uses natural language processing technology to analyze the speech content in the video and extract keywords and themes;

[0081] Furthermore, the emotion analysis unit performs emotion tendency analysis on the speech and pictures in the video based on the emotion classification model of LSTM, which is used to judge the emotional color of the video content; combines facial expression recognition technology to further refine the emotional expression of characters in the video.

[0082] In this embodiment, it should also be noted that the editing decision module further includes an editing point recommendation unit and a dynamic editing strategy unit;

[0083] Furthermore, the editing point recommendation unit uses inter-frame difference analysis to identify the shot transition points in the video and provides preliminary suggestions for editing; combines the quantity and rhythm changes to optimize the editing point recommendation to make the editing more smooth and natural;

[0084] Furthermore, the dynamic editing strategy unit dynamically adjusts the editing strategy according to the video content and the result of sentiment analysis by applying a reinforcement learning algorithm. For fast-paced videos, fast editing is adopted, while for slow-paced videos, slow editing is used; an adaptive editing length is achieved, and the clip duration is automatically adjusted according to the importance of the video content.

[0085] In this embodiment, it should also be noted that the special effect generation preview module further includes a real-time style transfer unit and a special effect preview interaction unit;

[0086] Furthermore, the real-time style transfer unit uses a style transfer model to achieve real-time conversion of the video style, including converting ordinary videos into oil painting styles and anime styles;

[0087] Through this unit, users are supported to customize the style, and style transfer is performed by uploading style pictures;

[0088] Furthermore, the special effect preview interaction unit uses augmented reality technology to achieve real-time preview of special effects, and users can directly see the special effect results during the editing process; an interactive special effect adjustment interface is provided, and users can adjust the special effect parameters through dragging and scaling operations.

[0089] In this embodiment, it should also be noted that the audio processing module further includes an audio separation unit and a sound effect addition unit;

[0090] Furthermore, the audio separation unit applies independent component analysis or a U-Net network to separate audio signals, including separating background music from vocals;

[0091] Through this unit, the separation and processing of multi-channel audio are supported, improving the flexibility of audio editing;

[0092] Furthermore, the sound effect addition unit automatically recommends and adds suitable sound effects or background music according to the video content and the result of sentiment analysis; custom sound effects are generated using waveform superposition audio synthesis to meet personalized needs.

[0093] In this embodiment, it should also be noted that the dynamic transition special effect module further includes a transition recognition unit and a transition special effect generation unit;

[0094] Furthermore, the transition recognition unit analyzes the video content through a sequence-to-sequence model to automatically identify the positions suitable for adding transition special effects; according to the video rhythm and emotional changes, the type and duration of the transition special effects are dynamically adjusted;

[0095] Furthermore, the transition special effect generation unit applies computer graphics technology to generate rich transition special effects, including dissolve, push-pull, and rotation;

[0096] This unit supports users to customize transition effects and provides various effect templates and parameter adjustment options.

[0097] In this embodiment, it should also be noted that the subtitle generation and editing module further includes an automatic subtitle generation unit and a subtitle style editing unit;

[0098] Furthermore, the automatic subtitle generation unit uses the ASR speech recognition method to convert the speech content in the video into text and automatically generates subtitles;

[0099] This unit supports the generation of multi-language subtitles to meet internationalization requirements;

[0100] Furthermore, the subtitle style editing unit provides various subtitle style templates, and users can select appropriate subtitle styles according to the video style;

[0101] This unit supports the custom editing of subtitle position, font, and color attributes to improve the readability and aesthetics of subtitles.

[0102] In this embodiment, it should also be noted that the virtual interaction module further includes a virtual character generation unit and an interaction element addition unit;

[0103] Furthermore, the virtual character generation unit applies Maya modeling and animation technology to create virtual characters, including animated characters and virtual anchors;

[0104] This unit supports character motion capture and expression mapping to achieve interaction of virtual characters in a real environment;

[0105] Furthermore, the interaction element addition unit provides a rich library of interaction elements, including buttons, sliders, and pop-ups. Users can add them to the video to enhance interactivity; logical control and event response of interaction elements are implemented using programming interfaces.

[0106] In this embodiment, it should also be noted that the content distribution optimization module further includes a platform adaptation unit and a content optimization suggestion unit;

[0107] Furthermore, the platform adaptation unit automatically adjusts video resolution, frame rate, and encoding format parameters according to the specifications and requirements of different video platforms; provides a one-click publishing function and supports simultaneous publishing to multiple video platforms;

[0108] Furthermore, the content optimization suggestion unit uses hot topic mining to provide optimization suggestions for video content; adjusts content strategies in real time according to performance data such as video play volume, like count, and comment count to improve video exposure rate and user participation.

[0109] Embodiment 2

[0110] Please refer to Figure 2, in the actual application process, for the intelligent digital media content editing and special effect generation method of artificial intelligence based on the above system, specifically, it includes the following steps:

[0111] S1. Video content recognition and sentiment analysis:

[0112] The system first receives the input video file, conducts content recognition on the video frames through a convolutional neural network, and automatically classifies and labels the scene, character, and object elements in the video;

[0113] At the same time, it uses natural language processing technology to analyze the speech content in the video, and extracts keywords and themes;

[0114] Then, based on the LSTM-based sentiment classification model, it conducts sentiment tendency analysis on the speech and pictures in the video, judges the emotional color of the video content, and combines facial expression recognition technology to refine the emotional expression of the characters;

[0115] S2. Editing point recommendation and dynamic editing strategy formulation:

[0116] The system uses inter-frame difference analysis to identify the shot transition points in the video, provides preliminary suggestions for editing, and optimizes the editing point recommendation in combination with volume and rhythm changes;

[0117] According to the video content and sentiment analysis results, it applies a reinforcement learning algorithm to dynamically adjust the editing strategy, including using fast editing for fast-paced videos and slow editing for slow-paced videos to achieve an adaptive editing length;

[0118] S3. Real-time style transfer and special effect preview interaction:

[0119] The system uses a style transfer model to achieve real-time conversion of the video style. Users can choose to convert the video into an oil painting style, an anime style, or upload a custom style picture for transfer;

[0120] It uses augmented reality technology to achieve real-time preview of special effects. Users can directly see the special effect effects during the editing process and adjust the special effect parameters through drag and zoom operations;

[0121] S4. Audio separation and sound effect addition:

[0122] The system applies independent component analysis or a U-Net network to achieve separation of audio signals, including separating background music from human voices, and supporting separation and processing of multi-channel audio;

[0123] According to the video content and sentiment analysis results, the system automatically recommends and adds suitable sound effects or background music. Users can also use waveform superposition audio synthesis to generate custom sound effects;

[0124] S5. Transition recognition and transition special effect generation:

[0125] The system analyzes the video content through a sequence-to-sequence model, automatically identifies the positions suitable for adding transition effects, and dynamically adjusts the type and duration of the transition effects according to the video rhythm and emotional changes;

[0126] Apply computer graphics technology to generate rich transition effects, including dissolve, push-pull, rotation. Users can choose preset templates or customize the effect parameters;

[0127] S6. Automatic subtitle generation and subtitle style editing:

[0128] The system uses the ASR speech recognition method to convert the speech content in the video into text, automatically generates subtitles, and supports the generation of multi-language subtitles;

[0129] Provide a variety of subtitle style templates for users to choose from. Users can adjust the subtitle position, font, and color attributes according to the video style;

[0130] S7. Virtual character generation and interactive element addition:

[0131] The system applies modeling animation technology to create virtual characters, including animated characters and virtual anchors, supports character motion capture and expression mapping, and realizes the interaction between virtual characters and the real environment;

[0132] Provide a rich library of interactive elements, including buttons, sliders, pop-ups. Users can add them to the video to enhance interactivity and implement logical control and event response through programming interfaces;

[0133] S8. Platform adaptation and content optimization suggestions:

[0134] The system automatically adjusts the video resolution, frame rate, and encoding format parameters according to the specifications and requirements of different video platforms, provides a one-click publishing function, and supports publishing to multiple video platforms simultaneously;

[0135] Use popular topic mining technology to provide optimization suggestions for video content, and adjust the content strategy in real time according to the performance data of video views, likes, and comments to improve video exposure and user engagement.

[0136] Through the above steps, the artificial intelligence-based intelligent digital media content editing and special effect generation system of the present invention has many advantages compared with traditional methods, can greatly improve the efficiency and quality of editing and special effect generation, meet the personalized needs of users, lower the technical threshold, and promote the innovation and development of digital media content.

[0137] Although embodiments of the present invention have been shown and described, those of ordinary skill in the art will appreciate that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent digital media content editing and special effect generation system based on artificial intelligence, characterized in that, It includes a content understanding and analysis module, a clip decision-making module, a special effect generation and preview module, an audio processing module, a dynamic transition special effect module, a subtitle generation and editing module, a virtual interaction module, and a content distribution optimization module; The content understanding and analysis module is used for video content recognition and sentiment analysis; The clip decision-making module is used for clip point recommendation and providing dynamic clip strategies; The special effect generation and preview module is used for real-time style transfer and special effect preview interaction; The audio processing module is used for audio separation and sound effect addition; The dynamic transition special effect module is used for transition recognition and transition special effect generation; The subtitle generation and editing module is used for automatic subtitle generation and subtitle style editing; The virtual interaction module is used for virtual character generation and interaction element addition; The content distribution optimization module is used for platform adaptation and content optimization suggestions.

2. The artificial intelligence-based digital media content intelligent editing and special effect generation system according to claim 1, wherein: The content understanding and analysis module further includes a video content recognition unit and a sentiment analysis unit; The video content recognition unit is used to automatically classify and label the scene, character, and object elements in the video; The sentiment analysis unit is used to judge the sentiment color of the video content.

3. The artificial intelligence-based digital media content intelligent editing and special effect generation system according to claim 2, wherein: The clip decision-making module further includes a clip point recommendation unit and a dynamic clip strategy unit; The clip point recommendation unit is used to provide preliminary suggestions for editing; The dynamic clip strategy unit is used to dynamically adjust the clip strategy.

4. The artificial intelligence-based digital media content intelligent editing and special effect generation system according to claim 3, wherein: The special effect generation and preview module further includes a real-time style transfer unit and a special effect preview interaction unit; The real-time style transfer unit is used to realize real-time conversion of the video style; The special effect preview interaction unit is used to realize real-time preview of the special effects.

5. The artificial intelligence-based digital media content intelligent editing and special effect generation system according to claim 4, wherein: The audio processing module further includes an audio separation unit and a sound effect addition unit; The audio separation unit is used to realize separation of audio signals; The sound effect addition unit is used to recommend and add suitable sound effects or background music.

6. The artificial intelligence-based digital media content intelligent editing and special effect generation system according to claim 5, wherein: The dynamic transition special effect module further includes a transition recognition unit and a transition special effect generation unit; The transition recognition unit is used to automatically identify the positions suitable for adding transition special effects; The transition special effect generation unit is used to generate rich transition special effects.

7. The artificial intelligence-based digital media content intelligent editing and special effect generation system according to claim 6, wherein: The subtitle generation and editing module further includes an automatic subtitle generation unit and a subtitle style editing unit; The automatic subtitle generation unit is used to convert the speech content in the video into text; The subtitle style editing unit is used to provide multiple subtitle style templates.

8. The digital media content intelligent editing and special effects generation system based on artificial intelligence according to claim 7 is characterized by: The virtual interaction module also includes a virtual character generation unit and an interactive element adding unit; The virtual character generation unit is used to create a virtual character; The interactive element adding unit is used to provide an interactive element library.

9. The artificial intelligence-based digital media content intelligent editing and special effects generation system according to claim 8, characterized in that: The content distribution optimization module also includes a platform adaptation unit and a content optimization suggestion unit; The platform adaptation unit is used to automatically adjust video resolution, frame rate, and encoding format parameters; The content optimization suggestion unit is used to provide optimization suggestions for video content.

10. A method for intelligent editing and special effect generation of digital media content based on artificial intelligence, according to the system for intelligent editing and special effect generation of digital media content based on artificial intelligence described in any one of claims 1-9, characterized in that, The following steps are involved: S1. The system first receives the input video file, performs content recognition on the video frame through the convolutional neural network, automatically classifies and annotates the scenes, characters, and object elements in the video, and uses natural language processing technology to analyze the voice content in the video, extract keywords and topics, and analyze the emotional tendency of the voice and picture in the video based on the LSTM sentiment classification model to determine the emotional color of the video content, and combines facial expression recognition technology to refine the emotional expression of the characters; S2. Use inter-frame difference analysis to identify shot switching points in the video, provide preliminary suggestions for editing, and optimize editing point recommendations based on volume and rhythm changes. Based on the video content and sentiment analysis results, apply reinforcement learning algorithms to dynamically adjust editing strategies; S3. Use style transfer model to achieve real-time conversion of video style, and use augmented reality technology to achieve real-time preview of special effects. Users can directly see the special effects during the editing process and adjust the special effects parameters by dragging and zooming; S4. Apply independent component analysis or U-Net network to separate audio signals, support separation and processing of multi-channel audio, and automatically recommend and add appropriate sound effects or background music based on video content and sentiment analysis results; S5. Analyze video content through sequence-to-sequence model, automatically identify suitable locations for adding transition effects, dynamically adjust the type and duration of transition effects according to video rhythm and emotional changes, and apply computer graphics technology to generate rich transition effects; S6. Use ASR speech recognition method to convert the speech content in the video into text, automatically generate subtitles, support multi-language subtitle generation, and provide a variety of subtitle style templates for users to choose; S7. Apply modeling animation technology to create virtual characters, including animated characters and virtual anchors, support character motion capture and expression mapping, realize the interaction between virtual characters and the real environment, and provide a rich library of interactive elements, including buttons, sliders, and pop-ups; S8. Automatically adjust video resolution, frame rate, and encoding format parameters according to the specifications and requirements of different video platforms, provide one-click publishing function, support simultaneous publishing to multiple video platforms, use hot topic mining technology to provide optimization suggestions for video content, and adjust content strategies in real time according to video playback volume, likes, and comments performance data to improve video exposure and user engagement.

Images