Automatic video editing method and system for clip list recommendation

Abstract

The invention discloses an automatic video editing method and system for film list recommendation, and the method comprises the steps: obtaining film data, and carrying out the preprocessing of the film data; according to the preprocessed film data, generating a recommendation list and basic data of a film list by utilizing a preset large model and cue words; generating a copywriting corresponding to the film list data according to the recommendation list and the basic data, and selecting corresponding film resources and background music; and according to the copywriting, the film resource and the background music, performing video synthesis editing, and outputting an automatic edited video recommended by a film list. According to the method, the material combinatorial logic can be dynamically adjusted according to the sheet theme, deep mining and structured application are performed on the sheet recommendation data, and high-efficiency and large-scale sheet recommendation video generation is realized.

Description

Technical Field

[0001] The present invention relates to the field of video processing technology, and in particular to an automated video editing method and system for recommending a playlist. Background Art

[0002] With the surge in demand for personalized playlist recommendations on short video platforms (e.g., film and television commentary collections, themed mashups, etc.), automated video editing technology is gradually shifting from single-scene editing to intelligent production that integrates multiple materials. Current mainstream technologies focus on processing single film materials (e.g., shot segmentation and filter adaptation), but when generating content for playlist recommendations, they still rely on manual material screening, script writing, and manual editing and splicing, which is inefficient and difficult to scale. For example, Douyin's "Film and Television Mashup" template only supports fixed shot durations to match background music and cannot dynamically adjust the material combination logic based on the playlist theme. While automated tools commonly used by streamers on some websites (e.g., Douyin AI Editor) can generate basic clips, they lack the in-depth mining and structured application of playlist recommendation data (e.g., user review keywords and film sentiment tags).

[0003] In summary, existing automated video editing technology still relies on manual screening of materials, script writing, and manual editing and splicing when generating content for playlist recommendations. This is inefficient and difficult to scale; therefore, the existing technology needs to be improved. Summary of the Invention

[0004] The technical problem to be solved by the present invention is that, in response to the defects of the existing technology, the present invention provides an automated video editing method and system for playlist recommendation, so as to solve the problem that the existing automated video editing technology still relies on manual screening of materials, script writing and manual editing and splicing when generating playlist-oriented recommendation content, which is inefficient and difficult to scale.

[0005] The technical solutions adopted by the present invention to solve the technical problems are as follows: In a first aspect, the present invention provides an automated video editing method for playlist recommendation, comprising: Acquiring video data and preprocessing the video data; Based on the pre-processed film data, a preset large model and prompt words are used to generate a recommended list and basic data for the film list; Generate text corresponding to the film list data based on the recommendation list and basic data, and select corresponding film resources and background music; Video synthesis and editing are performed according to the text, the film resources and the background music, and an automated editing video recommended by the playlist is output.

[0006] In one implementation, obtaining movie data and preprocessing the movie data include: Obtaining the film data from a preset media website; wherein the film data includes: film rating, film introduction, actor information, film review, and viewing experience; The film data is cleaned, films that do not meet the conditions are filtered out, and films of the same type are put into the same data set to obtain the preprocessed film data.

[0007] In one implementation, the method of generating a recommendation list and basic data of a film list based on the pre-processed film data using a preset large model and prompt words includes: Inputting the pre-processed video data into the preset macro model; Generating the prompt words according to the preferred tags, scene descriptions, number of films, and themes / types; Based on the preset large model, the prompt words are used to generate a recommendation list including the names of the film list and basic data including the film and background music.

[0008] In one implementation, generating text corresponding to the playlist data based on the recommendation list and basic data includes: The recommendation list and the basic data are input into the preset large model, and the corresponding introduction and commentary of the film is generated based on the editing operation recommended by the film list, and a fixed sentence pattern is added after the introduction and commentary to obtain the text corresponding to the film list data.

[0009] In one implementation, selecting corresponding movie resources and background music includes: Downloading the film resources corresponding to each film list from the media resource library corresponding to the basic data, and obtaining the introduction commentary, adding other prompt words to the introduction commentary, and selecting a film clip as the screen content when the introduction commentary is explained based on the preset large model; The music corresponding to each playlist is downloaded from the media resource library corresponding to the basic data, and the key parts of the music are selected based on the preset large model, and the key parts of the music are used as the background music for the key shots of the video recommended by the playlist.

[0010] In one implementation, the video synthesis and editing based on the text, the film resources, and the background music, and outputting the automatically edited video recommended by the playlist, includes: According to the text, the film resources and the background music, a film list recommendation structure of a preset editing template is used to perform video synthesis and editing, and an automated editing video recommended by the film list is output.

[0011] In one implementation, the video synthesis and editing is performed based on the text, the film resources, and the background music using a playlist recommendation structure of a preset editing template, and outputting an automated editing video recommended by the playlist, including: Set the first emoticon package at the beginning of the video recommended in the playlist, and set the greeting and author introduction information; According to the film resources, the corresponding film highlights are set in the main part of the video recommended in the playlist, and the MP3 version of the introduction and commentary of the copywriting, the corresponding text information and the background music for the key shots are configured; Set a second emoticon package at the end of the video recommended in the playlist, and set a preset ending word; Output the automatically edited video recommended by the playlist.

[0012] In a second aspect, the present invention provides an automated video editing system for recommending playlists, comprising: A data acquisition module, used to acquire film data and pre-process the film data; The recommendation list generation module is used to generate a recommendation list and basic data for the film list based on the pre-processed film data using a preset large model and prompt words; The video material module is used to generate the text corresponding to the film list data according to the recommendation list and basic data, and select the corresponding film resources and background music; The synthesis and editing module is used to perform video synthesis and editing based on the text, the film resources and the background music, and output the automatically edited video recommended by the playlist.

[0013] In a third aspect, the present invention provides a terminal comprising: a processor and a memory, wherein the memory stores an automated video editing program recommended by a playlist, and when the automated video editing program recommended by a playlist is executed by the processor, it is used to implement the operation of the automated video editing method recommended by a playlist as described in the first aspect.

[0014] In a fourth aspect, the present invention further provides a medium, which is a computer-readable storage medium, storing an automated video editing program for film list recommendations, and when the automated video editing program for film list recommendations is executed by a processor, is used to implement the operation of the automated video editing method for film list recommendations as described in the first aspect.

[0015] The present invention adopts the above technical solution to achieve the following effects: The present invention obtains and preprocesses film data; based on the preprocessed film data, it uses a preset large model and prompt words to generate a recommended list and basic data for the playlist; then, based on the recommended list and basic data, it generates a text corresponding to the playlist data and selects the corresponding film resources and background music; then, based on the text, film resources, and background music, it synthesizes and edits the video, outputting an automatically edited video of the playlist recommendations. The present invention dynamically adjusts the material combination logic based on the playlist theme, deeply mines and structures the playlist recommendation data, and achieves efficient and scalable generation of playlist recommendation videos. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the structures shown in these drawings without paying any creative work.

[0017] Figure 1 It is a flow chart of the automatic video editing method for film list recommendation in the present invention.

[0018] Figure 2 It is a functional principle diagram of a terminal in one implementation of the present invention.

[0019] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings and in conjunction with the embodiments. DETAILED DESCRIPTION

[0020] In order to make the purpose, technical solutions and advantages of the present invention more clear and distinct, the present invention is further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0021] Exemplary Methods With the surge in demand for personalized playlist recommendations on short video platforms (e.g., film and television commentary collections, themed mashups, etc.), automated video editing technology is gradually shifting from single-scene editing to intelligent production that integrates multiple materials. Current mainstream technologies focus on processing single film materials (e.g., shot segmentation and filter adaptation), but when generating content for playlist recommendations, they still rely on manual material screening, script writing, and manual editing and splicing, which is inefficient and difficult to scale. For example, Douyin's "Film and Television Mashup" template only supports fixed shot durations to match background music and cannot dynamically adjust the material combination logic based on the playlist theme. While automated tools commonly used by streamers on some websites (e.g., Douyin AI Editor) can generate basic clips, they lack the in-depth mining and structured application of playlist recommendation data (e.g., user review keywords and film sentiment tags).

[0022] In summary, existing automated video editing technology still relies on manual screening of materials, script writing, and manual editing and splicing when generating content for playlist recommendations. This is inefficient and difficult to scale; therefore, the existing technology needs to be improved.

[0023] In response to the above technical problems, an embodiment of the present invention provides an automated video editing method for playlist recommendations, the method comprising: obtaining film data and preprocessing the film data; generating a playlist recommendation list and basic data based on the preprocessed film data using a preset large model and prompt words; generating a text corresponding to the playlist data based on the recommendation list and basic data, and selecting corresponding film resources and background music; performing video synthesis and editing based on the text, the film resources, and the background music, and outputting an automated edited video recommended by the playlist. The present invention can dynamically adjust the material combination logic according to the playlist theme, conduct in-depth mining and structured application of the playlist recommendation data, and achieve efficient and large-scale playlist recommendation video generation.

[0024] like Figure 1 As shown, an embodiment of the present invention provides an automated video editing method for playlist recommendation, comprising the following steps: Step S100: Acquire video data and pre-process the video data.

[0025] In this embodiment, in order to implement the automated video editing method for playlist recommendation, an automated video editing system for playlist recommendation is proposed. Its technical innovations and advantages are as follows: 1) Dynamic playlist generation and data optimization: In this embodiment, a multi-level data cleaning pipeline is designed, combining film review sentiment analysis (for example, the BERT model) with user behavior profiling, dynamically expanding film tags (for example, "suspense index" and "tear point density") to improve the theme matching degree of the film list.

[0026] 2) Build a prompt word optimization engine to automatically generate fine-grained instructions based on the type of movie list (for example, "Generate a list of brain-burning science fiction movies that include the concept of 'time loop,' focusing on movies that mention 'reversal' in user reviews") to guide the large model to output a structured recommendation list.

[0027] 3) End-to-end efficient processing architecture: Adopt a distributed task scheduling framework: integrate crawlers, large model reasoning, material downloading and editing engines, optimize resource allocation through priority queues (for example, the urgency of the film list and user level), and reduce end-to-end processing latency.

[0028] In this embodiment, the crawler technology of the above system is used to obtain film data from third-party media websites (for example, free movie viewing and material platforms), and the film data (metadata) are automatically organized to obtain an organized film data set.

[0029] Specifically, in one implementation of this embodiment, step S100 includes the following steps: Step S101, obtaining the film data from a preset media website; wherein the film data includes: film rating, film introduction, actor information, film review and viewing experience; Step S102 , cleaning the film data, filtering out films that do not meet the conditions, and putting films of the same type into the same data set to obtain the pre-processed film data.

[0030] Since third-party media websites (for example, Mougua Film and Television, Videvo material platform) have a large amount of film introduction data, including film ratings, introductions, actor information and other content, more importantly, they have a large number of moviegoers' reviews and viewing experiences of the films. These data can serve as basic data to support film list recommendations. Therefore, in this embodiment, based on the crawler function of the above-mentioned system, film data including film ratings, film introductions, actor information, film reviews and viewing experiences are obtained from these third-party media websites. After obtaining a large amount of film data, the multi-level data cleaning pipeline of the above-mentioned system is used to enter the corresponding media resource library for data sorting, filter out films that do not meet the conditions (for example, films that do not conform to the user's viewing habits), and put the filtered films of the same type into the same data set to obtain the pre-processed film data.

[0031] like Figure 1 As shown, an embodiment of the present invention provides an automated video editing method for playlist recommendation, comprising the following steps: Step S200: Generate a recommended list and basic data of a film list based on the pre-processed film data using a preset large model and prompt words.

[0032] In this embodiment, after data is sorted through a multi-level data cleaning pipeline, the sorted number of films is input into a preset large model, and some pre-set prompt words are used to allow the large model to generate a recommendation list of high-quality film lists and basic data based on the film information. The recommendation list includes the film list name, and the basic data includes the film, background music, etc.

[0033] Specifically, in one implementation of this embodiment, step S200 includes the following steps: Step S201, inputting the pre-processed film data into the preset macro model; Step S202, generating the prompt word according to the preference tag, scene description, number of films, and theme / type; Step S203: Based on the preset large model, the prompt words are used to generate a recommendation list including the movie list names and basic data including the movies and background music.

[0034] In this embodiment, sentiment analysis of film reviews (e.g., BERT model) and user behavior profiling are combined to dynamically expand film tags (e.g., "suspense index," "tear density," etc.) to improve the theme matching of the film list, thereby generating a recommendation list containing the film list name and basic data including the film and background music.

[0035] As an example, when selecting a large model, you can use a generative large language model (for example, the BERT model, the GPT series model) or a hybrid recommendation framework (for example, the diffusion forced framework of SkyReels-V2). These large models support text generation and multimodal input; after selecting these large models, train them through fine-tuning.

[0036] During the training phase of the preset large model, based on the above-organized film data, user behavior data and film metadata (for example, genre, director, and actor) are used to construct a multi-dimensional feature vector. Then, contextual information (for example, viewing time and device type) is integrated to improve the dynamic recommendation effect. Finally, semantic understanding capabilities are enhanced through structured representation of film content (for example, shot composition and character expressions).

[0037] In this embodiment, the prompt word design process first establishes a basic prompt word template based on the user's historical preferences (preferred tags) and the current scene (scene description). A recommended number of films matching the theme / genre are then recommended. Reinforcement learning is then used to dynamically adjust the prompt word weights to improve the relevance of recommendations. Finally, the candidate set is retrieved in real time using an Elasticsearch-indexed film library to feed into the generation model, ensuring timely results. This embodiment utilizes a prompt word optimization engine to automatically generate fine-grained instructions based on the type of playlist (for example, "Generate a list of mind-bending sci-fi films that include the concept of 'time loop,' focusing on films that mention 'reversal' in user reviews"), guiding the large model to output a structured recommendation list.

[0038] In the process of generating a recommendation list containing the names of the film list and the basic data of the film and background music, the model outputs JSON format data containing the film name, recommendation reason, and matching score (range 0-1) to generate the recommendation list; then, the model automatically extracts film metadata (for example, duration, release year, award record), emotional tags (for example, "suspenseful and brain-burning", "healing") and similar film associations to generate basic data.

[0039] like Figure 1 As shown, an embodiment of the present invention provides an automated video editing method for playlist recommendation, comprising the following steps: Step S300: Generate text corresponding to the film list data based on the recommendation list and basic data, and select corresponding film resources and background music.

[0040] In this embodiment, after using the prompt words to generate a recommendation list containing the names of the playlists and the basic data containing the movies and background music, based on the above system, the text corresponding to the playlist data is automatically generated according to the recommendation list and the basic data, and the movie resources in the recommendation list are automatically downloaded from the corresponding media resource library, and the music corresponding to each playlist is automatically downloaded from the corresponding media resource library, so as to obtain the materials required for subsequent editing.

[0041] Specifically, in one implementation of this embodiment, step S300 includes the following steps: Step S301: Input the recommendation list and the basic data into the preset macro model, generate corresponding introduction and commentary of the film based on the editing operation recommended by the list, and add a fixed sentence pattern after the introduction and commentary to obtain the copy corresponding to the list data; Step S302: Download the film resources corresponding to each film list from the media resource library corresponding to the basic data, obtain the introduction and commentary, add other prompt words to the introduction and commentary, and select a film clip as the screen content for the introduction and commentary based on the preset macro model; Step S303: Download the music corresponding to each playlist from the media resource library corresponding to the basic data, select the key parts of the music based on the preset large model, and use the key parts of the music as the background music for the key shots of the video recommended by the playlist.

[0042] In this embodiment, after obtaining the playlist data (including the recommended list of playlist names and the basic data of the movies and background music), the subsequent operations are mainly divided into three steps: Big model copywriting generation: After inputting the film data of the playlist into the big model, tell the big model that the current operation is the editing recommended by the playlist, and let the big model generate some introduction and commentary containing the film. After obtaining the introduction and commentary, add some fixed sentence patterns and convert it into MP3 format (text-to-speech MP3), and add it as background music for the finished video.

[0043] Download movies and select shots (AI shot selection): After obtaining the movies included in the recommended list, you can download the corresponding movie resources in the media resource library, obtain the introduction and commentary generated by the big model in the previous step, and re-enter the big model with some other prompts, so that the big model can select some movie clips as the screen content for the introduction and commentary.

[0044] Download music and select the climax part (AI music climax excerpt): After obtaining the background music to be used in the playlist recommendation, download the music from the media resource database and let the big model select the climax part of the song. It can be used as the background music for the key shots of the finished video recommended by the playlist.

[0045] like Figure 1 As shown, an embodiment of the present invention provides an automated video editing method for playlist recommendation, comprising the following steps: Step S400: perform video synthesis and editing based on the text, the film resources, and the background music, and output an automatically edited video recommended by the playlist.

[0046] In this embodiment, after the copy of the playlist, AI shot selection, and AI music climax excerpts are automatically generated through the large model, video synthesis and editing can be performed based on the copy, AI-selected film resources, and background music. During the video synthesis and editing process, resource allocation can be optimized through priority queues (for example, the urgency of the playlist, user level), thereby reducing end-to-end processing delays.

[0047] Specifically, in one implementation of this embodiment, step S400 includes the following steps: Step S401, based on the text, the film resources and the background music, a video synthesis and editing is performed using a playlist recommendation structure of a preset editing template, and an automated editing video recommended by the playlist is output.

[0048] Specifically, in one implementation of this embodiment, step S401 includes: setting a first emoticon package at the beginning of the video recommended in the playlist, and setting a greeting and author introduction information; according to the film resources, setting corresponding high-energy shots of the film in the main part of the video recommended in the playlist, and configuring the mp3 version of the introduction commentary of the copy, the corresponding text information and the background music for the key shots; setting a second emoticon package at the end of the video recommended in the playlist, and setting a preset ending word; outputting the automatically edited video recommended by the playlist.

[0049] In this embodiment, based on a large model, through the semantic-visual-audio ternary alignment scheme and the dynamic resource orchestration algorithm, combined with GPU accelerated video preprocessing, efficient video synthesis and editing are performed, thereby realizing the automated production of high-quality videos and accurately outputting the automated editing videos recommended by the playlist.

[0050] As an example, in this embodiment, the steps of assembling a mashup video are as follows: 1. After obtaining all the required video materials, synthesize the materials according to the recommended structure of the playlist.

[0051] 2. The opening screen is an emoticon package of about 5 seconds, and the sound is a greeting + introduction to the author's information.

[0052] 3. The main part includes some high-energy shots of the film, and the background sound is the MP3 version of the introduction commentary given by the large model. The introduction commentary of each film is about 40 seconds. When entering the film introduction, some text information will be displayed on the screen in the first few seconds to help the audience obtain film information, such as the film name, actors, etc.

[0053] 4. The ending part is an emoticon package of about 3 seconds, which is also a relatively fixed ending word, for example: See you next time! 5. Add video shot switching and background music fade-in and fade-out, etc.

[0054] This embodiment integrates the capabilities of the large model to achieve a series of automated editing functions for film list recommendations, including film list generation, sound balancing, video animation, sound track analysis, text-to-speech, AI voice-over, shot segmentation, automatic subtitles, picture optical flow analysis, and picture star recognition.

[0055] This embodiment achieves the following technical effects through the above technical solution: This embodiment obtains and preprocesses film data; based on the preprocessed film data, it uses a preset large model and prompt words to generate a recommended list and basic data for the playlist; then, based on the recommended list and basic data, it generates a text corresponding to the playlist data and selects the corresponding film resources and background music; then, based on the text, film resources, and background music, it synthesizes and edits the video, outputting an automatically edited video of the playlist recommendation. This embodiment dynamically adjusts the material combination logic based on the playlist theme, conducts in-depth mining and structured application of the playlist recommendation data, and achieves efficient and scalable playlist recommendation video generation.

[0056] Exemplary devices Based on the above embodiment, the present invention further provides an automated video editing system for recommending playlists, comprising: A data acquisition module, used to acquire film data and pre-process the film data; The recommendation list generation module is used to generate a recommendation list and basic data for the film list based on the pre-processed film data using a preset large model and prompt words; The video material module is used to generate the text corresponding to the film list data according to the recommendation list and basic data, and select the corresponding film resources and background music; The synthesis and editing module is used to perform video synthesis and editing based on the text, the film resources and the background music, and output the automatically edited video recommended by the playlist.

[0057] This embodiment achieves the following technical effects through the above technical solution: This embodiment obtains and preprocesses film data; based on the preprocessed film data, it uses a preset large model and prompt words to generate a recommended list and basic data for the playlist; then, based on the recommended list and basic data, it generates a text corresponding to the playlist data and selects the corresponding film resources and background music; then, based on the text, film resources, and background music, it synthesizes and edits the video, outputting an automatically edited video of the playlist recommendation. This embodiment dynamically adjusts the material combination logic based on the playlist theme, conducts in-depth mining and structured application of the playlist recommendation data, and achieves efficient and scalable playlist recommendation video generation.

[0058] Based on the above embodiment, the present invention further provides a terminal, whose principle block diagram can be shown as follows: Figure 2 shown.

[0059] The terminal includes: a processor, memory, interface, display screen and communication module connected via a system bus; wherein the processor of the terminal is used to provide computing and control capabilities; the memory of the terminal includes a storage medium and an internal memory; the storage medium stores an operating system and computer programs; the internal memory provides an environment for the operation of the operating system and computer programs in the storage medium; the interface is used to connect to external devices; the display screen is used to display corresponding information; and the communication module is used to communicate with a cloud server or other devices.

[0060] When the computer program is executed by a processor, it is used to implement the operation of the automatic video editing method for playlist recommendation.

[0061] It will be understood by those skilled in the art that Figure 2 The principle block diagram shown in the figure is only a block diagram of a partial structure related to the solution of the present invention, and does not constitute a limitation on the terminal to which the solution of the present invention is applied. The specific terminal may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0062] In one embodiment, a terminal is provided, which includes: a processor and a memory, wherein the memory stores an automated video editing program recommended by a playlist, and when the automated video editing program recommended by a playlist is executed by the processor, it is used to implement the operation of the automated video editing method recommended by a playlist as described above.

[0063] In one embodiment, a storage medium is provided, wherein the storage medium stores an automated video editing program for film list recommendations, and when the automated video editing program for film list recommendations is executed by a processor, it is used to implement the operations of the automated video editing method for film list recommendations as described above.

[0064] Those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile storage medium. When executed, the computer program can include the processes in the above-described method embodiments. Any reference to memory, storage, database, or other media used in the various embodiments provided herein may include both non-volatile and volatile memory.

[0065] In summary, the present invention provides an automated video editing method and system for playlist recommendations, comprising: obtaining film data and preprocessing the film data; generating a playlist recommendation list and basic data based on the preprocessed film data using a preset large model and prompt words; generating a text corresponding to the playlist data based on the recommendation list and basic data, and selecting corresponding film resources and background music; synthesizing and editing the video based on the text, the film resources, and the background music, and outputting an automatically edited video of the playlist recommendations. The present invention can dynamically adjust the material combination logic according to the playlist theme, conduct in-depth mining and structured application of the playlist recommendation data, and achieve efficient and scalable playlist recommendation video generation.

[0066] It should be understood that the application of the present invention is not limited to the above examples. For those skilled in the art, improvements or changes can be made based on the above description. All these improvements and changes should fall within the scope of protection of the claims attached to the present invention.

Claims

1. An automated video editing method for playlist recommendation, characterized in that: include: Acquiring video data and preprocessing the video data; Based on the pre-processed film data, a preset large model and prompt words are used to generate a recommended list and basic data for the film list; Generate text corresponding to the film list data based on the recommendation list and basic data, and select corresponding film resources and background music; Video synthesis and editing are performed according to the text, the film resources and the background music, and an automated editing video recommended by the playlist is output.

2. The automatic video editing method for playlist recommendation according to claim 1, characterized in that: Acquiring video data and preprocessing the video data includes: Obtaining the film data from a preset media website; wherein the film data includes: film rating, film introduction, actor information, film review, and viewing experience; The film data is cleaned, films that do not meet the conditions are filtered out, and films of the same type are put into the same data set to obtain the preprocessed film data.

3. The automatic video editing method for playlist recommendation according to claim 1, characterized in that: The method of generating a recommendation list and basic data of a film list based on the pre-processed film data using a preset large model and prompt words includes: Inputting the pre-processed video data into the preset macro model; Generating the prompt words according to the preferred tags, scene descriptions, number of films, and themes / types; Based on the preset large model, the prompt words are used to generate a recommendation list including the names of the film list and basic data including the film and background music.

4. The automatic video editing method for playlist recommendation according to claim 1, characterized in that: Generating the copy corresponding to the film list data according to the recommendation list and basic data includes: The recommendation list and the basic data are input into the preset large model, and the corresponding introduction and commentary of the film is generated based on the editing operation recommended by the film list, and a fixed sentence pattern is added after the introduction and commentary to obtain the text corresponding to the film list data.

5. The automatic video editing method for playlist recommendation according to claim 4, characterized in that: The selected corresponding movie resources and background music include: Downloading the film resources corresponding to each film list from the media resource library corresponding to the basic data, and obtaining the introduction commentary, adding other prompt words to the introduction commentary, and selecting a film clip as the screen content when the introduction commentary is explained based on the preset large model; The music corresponding to each playlist is downloaded from the media resource library corresponding to the basic data, and the key parts of the music are selected based on the preset large model, and the key parts of the music are used as the background music for the key shots of the video recommended by the playlist.

6. The automatic video editing method for playlist recommendation according to claim 1, characterized in that: The video synthesis and editing is performed according to the text, the film resources and the background music, and the automatic editing video recommended by the playlist is output, including: According to the text, the film resources and the background music, a film list recommendation structure of a preset editing template is used to perform video synthesis and editing, and an automated editing video recommended by the film list is output.

7. The automatic video editing method for playlist recommendation according to claim 6, characterized in that: The step of synthesizing and editing the video using a playlist recommendation structure of a preset editing template according to the text, the film resources, and the background music, and outputting the automatically edited video recommended by the playlist, includes: Set the first emoticon package at the beginning of the video recommended in the playlist, and set the greeting and author introduction information; According to the film resources, the corresponding film highlights are set in the main part of the video recommended in the playlist, and the MP3 version of the introduction and commentary of the copywriting, the corresponding text information and the background music for the key shots are configured; Set a second emoticon package at the end of the video recommended in the playlist, and set a preset ending word; Output the automatically edited video recommended by the playlist.

8. An automated video editing system for playlist recommendation, characterized in that: include: A data acquisition module, used to acquire film data and pre-process the film data; The recommendation list generation module is used to generate a recommendation list and basic data for the film list based on the pre-processed film data using a preset large model and prompt words; The video material module is used to generate the text corresponding to the film list data according to the recommendation list and basic data, and select the corresponding film resources and background music; The synthesis and editing module is used to perform video synthesis and editing based on the text, the film resources and the background music, and output the automatically edited video recommended by the playlist.

9. A terminal, characterized in that: include: A processor and a memory, wherein the memory stores an automated video editing program for film list recommendations, and when the automated video editing program for film list recommendations is executed by the processor, it is used to implement the operation of the automated video editing method for film list recommendations as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores an automated video editing program for film list recommendations, which, when executed by a processor, is used to implement the operation of the automated video editing method for film list recommendations as described in any one of claims 1-7.

Images