Video clip tracing knowledge base construction method and device

Abstract

The embodiment of the invention discloses a video clip tracing knowledge base construction method and device. The method comprises the following steps: acquiring video editing data in real time; performing data structuring processing on the video clip data to obtain structured video clip data; data integration analysis processing is conducted on the structured video editing data, knowledge graph data is generated, and the knowledge graph data comprises at least two nodes, the incidence relation between the at least two nodes and node attributes; and storing the structured video clip data and the knowledge graph data in a video clip tracing knowledge base according to a preset data structure. The video editing data in the editing process is recorded into the video editing tracing knowledge base in real time, so that the editing history can be comprehensively and efficiently traced, the searching time of a user can be effectively saved, the user does not need to spend a lot of time for searching, the quality and efficiency of video production are improved, problem positioning and repairing can be carried out, and the user experience is improved. And the security and compliance use of the data can also be ensured.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of Internet technology, and specifically to a method, apparatus, computing device, computer storage medium, and computer program product for constructing a video clip tracing knowledge base. Background Art

[0002] With the increasing popularity of video creation, video editing has become a crucial aspect of both individual and team creation. However, traditional video editing management methods suffer from the difficulty of tracing the editing process. Specifically, user operations during the editing process (such as applying effects and adjusting timelines) lack systematic records, making it difficult to trace back. Summary of the Invention

[0003] In view of the above problems, the present application is proposed to provide a method, apparatus, computing device, computer storage medium and computer program product for constructing a video clip tracing knowledge base that overcomes the above problems or at least partially solves the above problems.

[0004] According to one aspect of an embodiment of the present application, a method for constructing a video clip tracing knowledge base is provided, comprising:

[0005] Get video clip data in real time;

[0006] Performing data structuring processing on the video clip data to obtain structured video clip data;

[0007] Performing data integration and analysis on structured video clip data to generate knowledge graph data, wherein the knowledge graph data includes: at least two nodes, an association relationship between at least two nodes, and node attributes;

[0008] The structured video clip data and knowledge graph data are stored in the video clip tracing knowledge base according to the preset data structure.

[0009] Furthermore, the video clip tracing knowledge base includes: a relational database and a graph database;

[0010] Storing the structured video clip data and the knowledge graph data in a video clip tracing knowledge base according to a preset data structure further includes:

[0011] Storing the structured video clip data in a relational database according to a first preset data structure;

[0012] The knowledge graph data is stored in the graph database according to the second preset data structure.

[0013] Furthermore, the video editing data includes: video material data, editing behavior data, and editing draft data;

[0014] Nodes include: video material node, editing behavior node, and draft version node;

[0015] The association relationship includes: the relationship between the video material and the editing behavior, and the relationship between the editing behavior and the draft version.

[0016] Furthermore, the method further includes: receiving a first tracing request, wherein the first tracing request includes: first tracing information;

[0017] Analyze the first traceability information, extract the first traceability keyword, and perform intent recognition processing on the first traceability information to determine the traceability intent;

[0018] Search the video clip tracing knowledge base according to the first tracing keyword and the tracing intention to obtain the first video clip tracing result;

[0019] Displaying the first video clip tracing result in a preset format in the user interface; and / or

[0020] The first video clip tracing result is processed by natural language conversion, and the conversion result is displayed on the user interface.

[0021] Furthermore, the tracing intention includes: editing behavior tracing, draft tracing, and video material tracing.

[0022] Furthermore, after displaying the first video clip tracing result in a preset form in the user interface; and / or performing natural language conversion processing on the first video clip tracing result and displaying the conversion result in the user interface, the method further includes:

[0023] Continue to obtain at least one second tracing request, wherein the second tracing request includes: second tracing information;

[0024] For any second tracing information, analyzing the second tracing information, extracting second tracing keywords, searching a video clip tracing knowledge base according to the second tracing keywords, and obtaining a second video clip tracing result;

[0025] Combining the first video clip tracing result and the second video clip tracing result, and displaying the combined processing result in a preset form in the user interface; and / or

[0026] A natural language conversion process is performed on the first video clip tracing result and the at least one second video clip tracing result, and the conversion result is displayed on a user interface.

[0027] Furthermore, the method further comprises: receiving a draft comparison request, wherein the draft comparison request includes an identifier of a draft to be compared;

[0028] According to the draft identifier to be compared, analyze and compare the corresponding video drafts in the video clip traceability knowledge base to generate draft differences;

[0029] Display the draft differences in a preset form in the user interface.

[0030] Furthermore, the method further includes: if a new video clip request is detected, retrieve the video clip traceability knowledge base according to the object identifier, determine the historical video clip data and object preference data, and recommend clip elements according to the historical video clip data and object preference data, where the historical video clip data includes: historical structured video clip data.

[0031] According to another aspect of the embodiments of the present application, there is provided a video clip traceability knowledge base construction device, including:

[0032] A data acquisition module, adapted to acquire video clip data in real time;

[0033] A data structuring processing module, adapted to perform data structuring processing on the video clip data to obtain structured video clip data;

[0034] A data integration and analysis processing module, adapted to perform data integration and analysis processing on the structured video clip data to generate knowledge graph data, where the knowledge graph data includes: at least two nodes, the association relationships between at least two nodes, and node attributes;

[0035] A video clip traceability knowledge base, adapted to store the structured video clip data and the knowledge graph data according to a preset data structure.

[0036] According to yet another aspect of the embodiments of the present application, there is provided a computing device, including: a processor, a memory, a communication interface, and a communication bus, and the processor, the memory, and the communication interface complete communication with each other through the communication bus;

[0037] The memory is used to store at least one executable instruction, and the executable instruction causes the processor to perform the operations corresponding to the above video clip traceability knowledge base construction method.

[0038] According to still another aspect of the embodiments of the present application, there is provided a computer storage medium, in which at least one executable instruction is stored, and the executable instruction causes the processor to perform the operations corresponding to the above video clip traceability knowledge base construction method.

[0039] According to still another aspect of the embodiments of the present application, there is provided a computer program product, including at least one executable instruction, and the executable instruction causes the processor to perform the operations corresponding to the above video clip traceability knowledge base construction method.

[0040] According to the video clip traceability knowledge base construction solution provided by the embodiments of the present application, by recording relevant video clip data during the clip process in real time into the video clip traceability knowledge base, it is convenient to comprehensively and efficiently trace the clip history, which can effectively save the user's search time and eliminate the need for the user to spend a large amount of time searching, such as searching for materials and adjusting effects. As a result, the quality and efficiency of video production can be improved. In addition, problem location and repair can be carried out, and the security and compliant use of data can also be ensured.

[0041] The above description is only an overview of the technical solutions of the embodiments of the present application. In order to be able to understand the technical means of the embodiments of the present application more clearly, it can be implemented in accordance with the content of the description. Moreover, in order to make the above and other purposes, features, and advantages of the embodiments of the present application more obvious and understandable, the following specifically describes the specific implementation manners of the embodiments of the present application. Brief Description of the Drawings

[0042] By reading the following detailed description of the preferred embodiments, various other advantages and benefits will become clear to those of ordinary skill in the art. The drawings are only for the purpose of showing the preferred embodiments and are not considered to be a limitation of the embodiments of the present application. Moreover, throughout the drawings, the same reference numerals are used to represent the same components. In the drawings:

[0043] Figure 1 A flowchart showing the process of a method for constructing a video clip traceability knowledge base according to an embodiment of the present application is shown;

[0044] Figure 2 A flowchart showing the process of a method for constructing a video clip traceability knowledge base according to another embodiment of the present application is shown;

[0045] Figure 3 A block diagram showing the structure of a video clip traceability knowledge base construction device according to an embodiment of the present application is shown;

[0046] Figure 4 A schematic diagram showing the structure of a computing device according to an embodiment of the present application is shown. Detailed Description of the Preferred Embodiments

[0047] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the drawings. Although the exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0048] First, the noun terms related to one or more embodiments of the present application are explained.

[0049] Knowledge Base: A collection of structured data stored locally by users, including all information related to video editing such as video materials, editing drafts, metadata (such as tags, descriptions, timeline information), editing behaviors (such as effect application, timeline adjustment), search records, user preferences, etc.

[0050] Editing Behavior: Operations performed by users during video editing, including but not limited to: setting editing points (in points, out points), applying effects (such as filters, transitions, subtitles), adjusting the timeline (such as clip order, duration), and adjusting audio (such as volume, sound effects, background music).

[0051] Editing Draft: Intermediate versions generated by users during the editing process, including information such as timeline configuration, effect application, clip order, and audio synchronization.

[0052] Traceability: The system can, according to user requirements, trace back editing behaviors, draft versions, material usage records, etc., and provide detailed context information.

[0053] With the popularization of video creation, video editing has become an important part of personal and team creation. However, the traditional video editing management method has the following problems:

[0054] 1. Difficulty in tracing editing behaviors: The operations of users during the editing process (such as effect application, timeline adjustment) lack systematic records, making it difficult to trace back and reuse.

[0055] 2. Chaotic material management: The number of video materials is huge, and there is a lack of effective classification and retrieval methods.

[0056] 3. Low efficiency: Users need to spend a lot of time searching for materials and adjusting effects, and there is a lack of intelligent auxiliary tools.

[0057] 4. Insufficient personalized support: Existing editing software cannot provide personalized suggestions based on users' historical behaviors and preferences.

[0058] Although existing video editing software (such as Adobe Premiere, Final Cut Pro) is powerful, it lacks in-depth analysis and utilization of the user's editing process. Therefore, there is an urgent need for a clip traceability solution based on a knowledge base to solve the above problems.

[0059] Figure 1 The flowchart of the method for constructing a video editing traceability knowledge base according to an embodiment of the present application is shown, as Figure 1 shown, the method includes the following steps:

[0060] Step S101, obtaining video editing data in real time.

[0061] Video editing data refers to data related to video editing. Sources include original video source files (such as MP4 and MOV formats), editing project files, and editing operation logs (such as user operation records and timeline modification history). This data can be collected through file import, API calls, or log capture.

[0062] The video editing data may include: video material data, editing behavior data, and editing draft data. Here, the video editing data is obtained in real time. The video editing data related to each editing operation will be obtained in time to facilitate the timely update of the constructed video editing traceability knowledge base.

[0063] It should be noted that the video editing data involved in the embodiments of the present application are all data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of relevant countries and regions, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0064] Step S102: performing data structuring processing on the video clip data to obtain structured video clip data.

[0065] Data structuring processing of video clip data refers to converting the original unstructured or semi-structured video clip data (such as video files, editing operation records, timeline markers, special effect parameters, etc.) into structured data with clear logical relationships and standardized formats for easy storage, retrieval and analysis.

[0066] By performing data structuring processing on the video clip data, corresponding structured video clip data can be obtained.

[0067] Step S103: perform data integration and analysis on the structured video clip data to generate knowledge graph data, wherein the knowledge graph data includes: at least two nodes, an association relationship between at least two nodes, and node attributes.

[0068] Specifically, after obtaining structured video clip data, data integration and analysis are conducted to construct knowledge graph data. For example, nodes can be defined based on data type. Once the nodes are determined, the relationships between nodes can be determined through data cross-references, operational logic deduction, and other methods, and the structured video clip data associated with the nodes can be used as node attributes. The defined nodes, relationships, and attributes are organized according to preset requirements to generate knowledge graph data, which fully presents the internal connections between video clip data.

[0069] Step S104: Store the structured video clip data and the knowledge graph data in a video clip tracing knowledge base according to a preset data structure.

[0070] To achieve the effective management and traceability of video clip data, the structured video clip data and knowledge graph data are stored in the video clip traceability knowledge base. In an alternative embodiment, the video clip traceability knowledge base includes: a relational database and a graph database; thus, the structured video clip data can be stored in the relational database according to a first preset data structure, and the relational database can be a SQLite or MySQL database, and the knowledge graph data can be stored in the graph database according to a second preset data structure, and the graph database can be a Neo4j database.

[0071] The video clip traceability knowledge base is used to store and manage all clip-related data of users, thereby enabling comprehensive traceability.

[0072] According to the method for constructing a video clip traceability knowledge base provided by an embodiment of the present application, by recording the video clip data related during the clip process in real time into the video clip traceability knowledge base, it is convenient to comprehensively and efficiently trace the clip history, which can effectively save the user's search time and does not require the user to spend a lot of time searching, for example, searching for materials, adjusting effects, etc., thereby improving the quality and efficiency of video production. In addition, problem location and repair can also be performed, and the security and compliant use of data can also be ensured.

[0073] Figure 2 The flowchart of the method for constructing a video clip traceability knowledge base according to another embodiment of the present application is shown, and this method is executed by the client, as Figure 2 shown, and this method includes the following steps:

[0074] Step S201, obtain video material data, clip behavior data, and clip draft data in real time.

[0075] Video material data refers to the descriptive data related to video materials in a video clip project. Among them, the video material data can include one or more of the following information: material ID, file name, path, label, duration, resolution, scene information, color distribution, object, etc.

[0076] Specifically, for a video, video frames are extracted from the video. Here, only key frames can be extracted, or video frames can be extracted at fixed time intervals. Then, the video frames of the video file are analyzed through computer vision technology to extract video metadata, such as metadata of scenes, objects, color distribution, etc. A deep learning model (such as ResNet, YOLO) is used to analyze the video content to generate labels, for example, travel videos, night scenes, close-ups of people, beaches, sunsets, etc.

[0077] For each extracted video frame, a deep learning-based scene classification model is used to determine the scene to which it belongs. For example, this scene classification model is pre-trained on a large-scale scene image dataset and can identify a variety of scene categories, such as beaches, city streets, forests, and offices. When a video frame is input into this scene classification model, the model outputs a probability value for each scene category and selects the category with the highest probability as the scene corresponding to the video frame.

[0078] Use object detection algorithms, such as the YOLO (You Only Look Once) family of algorithms, to detect and identify objects in video frames. Object detection algorithms can quickly and accurately identify various object categories in video frames, such as people, cars, and animals.

[0079] For example, use a deep learning model (such as ResNet, YOLO) to generate labels "beach", "sunset" and "people" for the video "Hawaii Trip.mp4".

[0080] Convert the video frame's color space from the common RGB color space to the HSV (hue, saturation, value) color space, which is more suitable for color analysis. Then, count the pixel distributions for each color channel (hue, saturation, value). This color distribution can be visualized by calculating a color histogram, which reflects the frequency of occurrence of different color values in an image. For images, the implementation is similar to that for videos and will not be detailed here. The acquisition of other data is not described here.

[0081] Audio analysis technology is used to analyze audio files and identify audio features such as background music and sound effects. For example, a pre-trained audio recognition model can be used to identify audio files and determine audio features such as background music and sound effects. The audio recognition model training process is as follows: a large number of audio files are obtained, audio features are extracted from the audio files, and the model can be trained based on the labeled data and audio features to obtain an audio recognition model.

[0082] For example, it uses audio feature extraction technology (such as MFCC) to analyze the audio file "Hawaii Trip.wav" and identify the background music as "Ukulele Song." Furthermore, it records usage data for each video clip, including: which projects the video clip is used in; its position on the timeline; and which effects are applied to the video clip.

[0083] Editing behavior data records a series of operational behavior information performed during the video editing process, including editing point settings (in point, out point), effect application (such as filters, transitions, subtitles), timeline adjustment (such as clip order and duration), audio adjustment (such as volume, sound effects, background music), etc. For example, the import, deletion, and sorting of materials; the cropping of clips, including determining the start and end points of cropping; the addition of transition effects, such as fade in and out, flash white, dissolve and other special effects; audio adjustment, such as increasing or decreasing the volume, audio mixing processing; and the addition and editing of subtitles, etc. For example, the corresponding editing behavior data can be obtained from the editing operation log, or the editing behavior data can be collected through the embedding technology.

[0084] Draft edit data is data related to video drafts that are saved periodically during the video editing process. Each editing operation generates a draft version, which records information such as timeline configuration, applied effects, and clip order. For example, you can obtain draft edit data from editing project files or collect it using tracking technology.

[0085] Video material data, editing behavior data, and editing draft data are obtained in real time. The video material data, editing behavior data, and editing draft data related to each editing operation will be obtained in a timely manner to facilitate timely updating of the constructed video editing traceability knowledge base.

[0086] Step S202: perform data structuring processing on the video material data, editing behavior data, and editing draft data to obtain video material structured data, editing behavior structured data, and editing draft structured data.

[0087] The collected video material data, editing behavior data, and editing draft data have diverse formats and complex structures. To achieve efficient storage and analysis, data structuring is required. For video material data, it can be organized according to a predetermined data structure to generate video material structured data. For editing behavior data, the type, execution time, and operation parameters of various editing operations are extracted based on timestamps and operation instructions, and organized into editing behavior structured data. For editing draft data, by analyzing the structure of the draft file, key information such as the material arrangement order and transition effects is extracted to construct editing draft structured data. Through this series of processing, the raw, unordered data is converted into a structured, standardized data format, facilitating subsequent data integration and analysis.

[0088] For example, for clip behavior data, the structured data example is as follows: A 'cross dissolve' transition is applied at timeline 00:10.

[0089] Step S203: Perform data integration analysis on the structured data of video materials, the structured data of editing behaviors, and the structured data of editing drafts to generate knowledge graph data. The knowledge graph data includes: at least two nodes, the association relationships between at least two nodes, and node attributes.

[0090] Specifically, after obtaining the structured data of video materials, the structured data of editing behaviors, and the structured data of editing drafts, conduct data integration analysis to construct knowledge graph data. Among them, the nodes include: video material nodes, editing behavior nodes, and draft version nodes; the association relationships include: the relationship between video materials and editing behaviors and the relationship between editing behaviors and draft versions.

[0091] First, define nodes according to the data types. Here, each video material is regarded as a video material node, and its attributes include detailed information such as the name, format, and resolution of the material; each type of editing behavior is defined as an editing behavior node, and the attributes cover the behavior ID, operation type, and operation details; each video draft version is defined as a draft version node, and the attributes include the draft ID, timeline configuration, and effect application.

[0092] Second, determine the association relationships between nodes. For example, determine the association relationships between nodes through ID matching, determine the association relationships between nodes through editing behavior logic, and determine the association relationships between nodes through effect application logic.

[0093] In the structured data of video materials, the structured data of editing behaviors, and the structured data of editing drafts, there are unique identifiers. By matching these IDs, the connections between different data can be established. For example, the video material ID on which the editing behavior acts will be recorded in the editing behavior data. Using this video material ID, the editing behavior node can be associated with the corresponding video material node, indicating that this editing behavior is carried out for a specific video material;

[0094] Establish the association between the video material node and the editing behavior node through the video material ID on which the editing behavior acts. For example, the splicing behavior means that multiple video materials are combined in a specific order. By analyzing the parameters of the splicing operation, the video material nodes participating in the splicing can be associated with this editing behavior node with an association relationship of "being spliced", and it can also be an association relationship of "being cropped", "applied with", "deleted".

[0095] The association relationship between the editing behavior node and the draft version node can be: generated, modified, etc. Here, it is only an example and has no restrictive effect.

[0096] The node attributes of the video material node include: material ID, label, path, etc.

[0097] The node properties of the clip behavior node include: behavior ID, operation type, operation details, etc.

[0098] The node properties of the draft version node include: draft ID, timeline configuration, effect application, etc.

[0099] Finally, the defined nodes and relationships are organized according to preset requirements to generate knowledge graph data, which fully presents the internal connections between video clip data.

[0100] Step S204: store the video material structured data, editing behavior structured data, and editing draft structured data in a relational database according to a first preset data structure; and store the knowledge graph data in a graph database according to a second preset data structure.

[0101] The video clip tracing knowledge base in this embodiment includes: a relational database and a graph database. Therefore, for structured video clip data, according to the database table structure design, the video material structured data, the editing behavior structured data, and the editing draft structured data are stored in corresponding tables respectively to ensure the standardization and integrity of the data. For example, the video material structured data can be stored in the video material table, the editing behavior structured data can be stored in the editing behavior table, and the editing draft structured data can be stored in the editing draft table. Among them, the relational database can be SQLite or MySQL.

[0102] Table structure design example:

[0103] Video material table: `Material ID`, `File Name`, `Path`, `Tag`, `Duration`, `Resolution`.

[0104] Clip action table: `action ID`, `user ID`, `operation type`, `operation time`, `operation details`.

[0105] Clip draft table: `Draft ID`, `Project ID`, `Timeline Configuration`, `Effects Applied`, `Creation Time`.

[0106] For knowledge graph data, use the storage interface provided by the knowledge graph construction tool to store the graph data in the form of a graph database to ensure that the graph's topological structure and association relationships are fully preserved. The graph database can be a Neo4j database.

[0107] This hierarchical storage approach facilitates traditional queries and statistical analysis of structured data while leveraging the knowledge graph to enable in-depth mining and tracing of video clip data relationships. The video clip tracing knowledge base is a local database, storing all data locally. Relevant data is managed with projects. For example, draft versions are stored in the video clip tracing knowledge base and associated with related projects.

[0108] Before step S204, an encryption algorithm (for example, an AES encryption algorithm) can also be used to encrypt the video material structured data, editing behavior structured data, editing draft structured data, and knowledge graph data, and store the encrypted video material structured data, editing behavior structured data, and editing draft structured data in a relational database; and store the encrypted knowledge graph data in a graph database, thereby ensuring data security and protecting user privacy.

[0109] Step S205: receiving a first tracing request, wherein the first tracing request includes: first tracing information.

[0110] Specifically, the client may provide a tracing page to the user. When the user has a clip tracing requirement, the user may send a first tracing request by asking a question in natural language on the tracing page. Alternatively, the tracing page may be a graphical interface, and the user may send the first tracing request by selecting a corresponding graphic, thereby receiving the first tracing request sent by the user. The first tracing request includes: first tracing information.

[0111] Taking natural language questions as an example, the first traceable information can be: "In which projects did I use 'beach.mp4' before?", "In which project did I use the 'cross dissolve' transition before?", "In which draft did I adjust the timeline before?", etc.

[0112] Step S206 : Analyze the first tracing information, extract the first tracing keyword, and perform intent recognition processing on the first tracing information to determine the tracing intent.

[0113] Specifically, a pre-trained language model (such as BERT or GPT) can be used to parse the first traceback information and extract first traceback keywords. For example, if the first traceback information is "In which project did I use the 'cross dissolve' transition?", the first traceback keywords extracted might be "cross dissolve" and "project"; if the first traceback information is "In which projects did I use 'beach.mp4'?", the first traceback keywords extracted might be "beach.mp4" and "project"; if the first traceback information is "In which draft did I adjust the timeline?", the first traceback keywords extracted might be "draft," "adjustment," and "timeline." The language model training process can be referenced in existing model training methods and will not be elaborated here.

[0114] For example, a pre-trained intent recognition model can be used to classify the first tracing information to determine the user's tracing intent, where tracing intent includes: editing behavior tracing, draft tracing, and video material tracing. Therefore, the first tracing information can be classified into types such as editing behavior tracing, draft tracing, and video material tracing. The intent recognition model can be trained based on labeled data. A large amount of question information for video editing tracing can be collected and labeled to obtain corresponding labeled data. For example, the labeled data can be editing behavior tracing, draft tracing, and video material tracing. The text features of the question information are extracted, and the model is trained based on the text features and the labeled data to obtain the intent recognition model.

[0115] For example, for the first tracing information: In which project did I use the 'cross dissolve' transition before?, the tracing intention can be determined as: editing behavior tracing.

[0116] Step S207 : searching a video clip tracing knowledge base according to the first tracing keyword and the tracing intention to obtain a first video clip tracing result.

[0117] After determining the first tracing keyword and tracing intention, the video clip tracing knowledge base can be searched according to the first tracing keyword and tracing intention, mainly by using a graph traversal algorithm to retrieve relevant records from the graph database, thereby obtaining the corresponding first video clip tracing result.

[0118] Continuing with the above example, when tracing back the editing behavior, it can be determined that the tracing result of the first video editing is 'Hawaii Trip'; when tracing back the draft, it can be determined that the tracing result of the first video editing is 'Draft V3'; when tracing back the video material, it can be determined that the tracing results of the first video editing are 'Hawaii Trip' and 'Bali Trip'. These are only examples and do not have any limiting effect.

[0119] Step S208: Display the first video clip tracing result in a preset form in the user interface; and / or, perform natural language conversion processing on the first video clip tracing result, and display the conversion result in the user interface.

[0120] In the user interface, the first video clip tracing results are accurately displayed in a preset format. This preset format can be a timeline, chart, or other visual format, making it easier for users to understand or select, improving the efficiency of users' information acquisition, for example, making it easier for users to select the best draft version.

[0121] Natural language conversion processing technology (for example, a large language model) can also be used to convert the first video clip tracing result. During the conversion process, the first tracing keyword can be combined for conversion. The conversion result is described in concise and clear sentences so that non-professional users can easily understand it. The conversion result is displayed on the user interface, making it convenient for users to view and understand the tracing result at any time, thereby providing users with a more friendly and efficient interactive experience.

[0122] For example, the conversion results can be: you used the 'Cross Dissolve' transition in the 'Hawaii Trip' project; you adjusted the timeline in 'Draft V3'; you used 'Beach.mp4' in the 'Hawaii Trip' and 'Bali Trip' projects.

[0123] Step S209: Continue to obtain at least one second tracing request, wherein the second tracing request includes: second tracing information.

[0124] After viewing the corresponding results, the user can continue to ask questions based on the above results, for example, the user continues to ask "at what time point the retro filter is applied?" The implementation process of this step is similar to step S205 and will not be repeated here.

[0125] When there are multiple second tracing requests, the multiple second tracing requests are sent successively, and after corresponding results are given, the next second tracing request is sent.

[0126] Step S210 : for any second tracing information, analyze the second tracing information, extract second tracing keywords, search the video clip tracing knowledge base according to the second tracing keywords, and obtain a second video clip tracing result.

[0127] The specific processing process is similar to step S206-step S207 and will not be repeated here.

[0128] Step S211, perform a combination process on the first video clip tracing result and the second video clip tracing result, and display the combination result in a preset form in the user interface; and / or, perform natural language conversion processing on the first video clip tracing result and at least one second video clip tracing result, and display the conversion result in the user interface.

[0129] By comprehensively considering various data information in the two tracing results, including but not limited to the time stamps of video clips, clip operation types (such as cropping, splicing, special effect addition, etc.), the sources of video segments involved, and relevant metadata, etc., through intelligent algorithms, these information are organically integrated to form a logically coherent and well-structured combination result. After completing the combination process, this combination result will be displayed in a preset form in the user interface.

[0130] and / or, use natural language conversion processing technology (for example, large language models) to perform conversion processing on the first video clip tracing result and at least one second video clip tracing result. During the conversion process, it can be combined with the first tracing keyword and the second tracing keyword for conversion. The conversion result is described in concise and clear statements, enabling non-professional users to easily understand, and the conversion result is displayed in the user interface, facilitating users to view and understand the tracing result at any time, thereby providing a more user-friendly and efficient interaction experience. For example, continuing with the above example, the result of performing natural language conversion processing on the first video clip tracing result and at least one second video clip tracing result can be: You added a 'vintage filter' at '00:23' in the 'Hawaii Trip' project.

[0131] In an optional implementation manner of the present application, the method further includes: receiving a draft comparison request, where the draft comparison request includes an identifier of the draft to be compared; according to the identifier of the draft to be compared, analyze and compare the corresponding video draft in the video clip tracing knowledge base to generate draft differences; and display the draft differences in a preset form in the user interface.

[0132] Specifically, during the video clip process, different versions of video drafts may be stored. Users may need to understand the differences between different versions of video drafts to optimize the clip process and improve video quality. Therefore, a draft comparison function can be provided. For example, a draft comparison button is displayed in the user interface. After the user clicks the draft comparison button, an input box or a draft version selection dropdown box can be displayed. When the user enters the identifier of the video draft to be compared in the input box, or selects the identifier of the video draft to be compared through the draft version selection dropdown box, it is regarded as sending a draft comparison request, where the draft comparison request includes an identifier of the draft to be compared. The draft identifier is the key to accurately screening and locating the target video draft.

[0133] Using the pending draft identifier, the corresponding video draft can be quickly located in the video editing traceability knowledge base. The relational database stores draft editing data, such as timeline configuration, effects applied, and clip order. Therefore, this information can be quickly retrieved using the pending draft identifier. Analysis and comparison of this draft editing data can identify differences between drafts, such as differences in clip order, duration, transition effects, timeline, subtitles, and audio adjustments. Draft differences are visualized in the user interface to facilitate user selection of the optimal version. For example, a visual interface uses side-by-side comparison, timeline highlighting, and pop-up notifications to intuitively present draft differences. For example, in side-by-side comparison mode, two video draft versions are displayed on the same screen, with differences highlighted in different colors or with special effects for quick identification. Timeline highlighting clearly marks the difference segments on the timeline, allowing users to click on the marker to view the specific differences. This helps creators quickly understand the changes between different drafts and make more informed creative decisions.

[0134] In addition, you can categorize and organize the detected draft differences into a draft difference report. The draft difference report covers details such as the addition, deletion, and replacement of clips, changes to transition effects, adjustments to subtitle text or style, and modifications to audio parameters.

[0135] In an optional embodiment of the present application, the method also includes: if a new video editing request is monitored, retrieving the video editing tracing knowledge base based on the object identifier, determining the historical video editing data and the object preference data, and recommending editing elements based on the historical video editing data and the object preference data, wherein the historical video editing data includes: historical structured video editing data.

[0136] Specifically, editing elements refer to the various elements and factors that play a key role in the final presentation effect of the video during the video editing process, mainly including video materials, effects and editing style.

[0137] In order to improve the editing effect of users, when there is a new video editing project that needs to be edited, the video editing tracing knowledge base is retrieved according to the object identifier. For example, the relational database is retrieved to determine the historical video editing data and the object preference data. Among them, the object preference data can be determined in advance based on the analysis of the historical video editing data and stored in the video editing tracing knowledge base, so as to be directly obtained from the video editing tracing knowledge base, or it can be obtained based on the analysis of the historical video editing data after monitoring the new video editing request.

[0138] Then, editing elements are recommended based on historical video editing data and object preference data, where historical video editing data includes: historical structured video editing data. For example, based on data related to historical editing behaviors, commonly used effects (such as "cross dissolve" transitions) can be recommended, and editing styles (such as "fast-paced editing") can be recommended based on user preferences. For example, collaborative filtering algorithms are used to analyze editing-related data, and relevant materials and effects (such as "beach.mp4" and "cross dissolve" transitions) are recommended to users. The graph traversal algorithm of the graph database is used to quickly find relevant video materials and effects and recommend them to users. Users often use "cross dissolve" transitions and "retro filters", and these effects can be automatically recommended in new editing projects.

[0139] In addition, you can also obtain the subject and tags of the new video editing project, query the material library based on the subject and tags, and recommend related video materials, such as recommending "beach.mp4" for "travel video".

[0140] Through the above recommendations, the above editing elements can be automatically applied in new video editing projects to complete part of the editing tasks, thereby improving editing efficiency.

[0141] Every operation performed by the user during the editing process will be recorded in real time and stored in the video editing traceability knowledge base. By building a comprehensive editing behavior recording and tracing mechanism, the following effects are achieved: users can query editing behaviors, draft versions, material usage records, etc. through natural language or graphical interfaces, and can trace the editing history more efficiently; the draft version of each editing operation is automatically recorded, supporting backtracking, comparison and reuse, and realizing convenient management of draft versions; through intelligent recommendations and automation functions, repetitive work is reduced and editing efficiency is improved; all data is stored locally to ensure privacy and security.

[0142] In addition, the video clips are tracked back to the knowledge base, and video clip data and user preferences are regularly updated to ensure the timeliness of the knowledge base. The transaction mechanism is used to ensure data consistency and avoid data loss or conflicts.

[0143] When users perform clip tracing, the submitted tracing information and click behavior information on the results are obtained, and the above information is recorded in the video clip tracing knowledge base for object preference analysis.

[0144] According to the video editing tracing knowledge base construction method provided by the embodiment of the present application, by recording the relevant video editing data in the editing process into the video editing tracing knowledge base in real time, it is convenient to comprehensively and efficiently trace the editing history, which can effectively save users' search time and eliminate the need for users to spend a lot of time searching, for example, searching for materials, adjusting effects, etc., thereby improving the quality and efficiency of video production. In addition, it can also locate and repair problems, and ensure the security and compliance of data use; editing elements are recommended through historical video editing data and object preference data, and the above-mentioned editing elements can be automatically applied in new video editing projects to complete part of the editing tasks, thereby improving editing efficiency; the draft version of each editing operation is automatically recorded, supporting backtracking, comparison and reuse, and realizing convenient management of draft versions.

[0145] Figure 3 FIG. 1 shows a structural block diagram of a device for constructing a video clip tracing knowledge base according to an embodiment of the present application. Figure 3 As shown, the device includes: a data acquisition module 301, a data structuring processing module 302, a data integration analysis processing module 303, and a video clip tracing knowledge base 304.

[0146] A data acquisition module 301 is adapted to acquire video clip data in real time;

[0147] The data structuring processing module 302 is adapted to perform data structuring processing on the video clip data to obtain structured video clip data;

[0148] The data integration analysis processing module 303 is adapted to perform data integration analysis on the structured video clip data to generate knowledge graph data, wherein the knowledge graph data includes: at least two nodes, an association relationship between at least two nodes, and node attributes;

[0149] The video clip tracing knowledge base 304 is suitable for storing structured video clip data and knowledge graph data according to a preset data structure.

[0150] Optionally, the video clip tracing knowledge base includes: a relational database and a graph database;

[0151] a relational database adapted to store structured video clip data according to a first preset data structure;

[0152] The graph database is suitable for storing knowledge graph data according to a second preset data structure.

[0153] Furthermore, the video editing data includes: video material data, editing behavior data, and editing draft data;

[0154] Nodes include: video material node, editing behavior node, and draft version node;

[0155] The association relationships include: the relationship between video materials and editing operations, and the relationship between editing operations and draft versions.

[0156] Furthermore, the device further includes: a tracing module, adapted to receive a first tracing request, where the first tracing request includes: first tracing information;

[0157] Analyze the first tracing information, extract first tracing keywords, and perform intention recognition processing on the first tracing information to determine the tracing intention;

[0158] Retrieve a video editing tracing knowledge base according to the first tracing keywords and the tracing intention to obtain a first video editing tracing result;

[0159] Display the first video editing tracing result in a preset form in the user interface; and / or

[0160] Perform natural language conversion processing on the first video editing tracing result and display the conversion result in the user interface.

[0161] Furthermore, the tracing intention includes: editing operation tracing, draft tracing, and video material tracing.

[0162] Furthermore, the tracing module is further adapted to: continue to obtain at least one second tracing request, where the second tracing request includes: second tracing information;

[0163] For any second tracing information, analyze the second tracing information, extract second tracing keywords, retrieve a video editing tracing knowledge base according to the second tracing keywords to obtain a second video editing tracing result;

[0164] Perform combination processing on the first video editing tracing result and the second video editing tracing result, and display the combination processing result in a preset form in the user interface; and / or

[0165] Perform natural language conversion processing on the first video editing tracing result and at least one second video editing tracing result, and display the conversion result in the user interface.

[0166] Furthermore, the device further includes: a draft comparison module, adapted to receive a draft comparison request, where the draft comparison request includes a draft identifier to be compared;

[0167] Analyze and compare the corresponding video drafts in the video editing tracing knowledge base according to the draft identifier to be compared to generate draft differences;

[0168] Display the draft differences in a preset form in the user interface.

[0169] Furthermore, the device also includes: a recommendation module, which is suitable for retrieving the video clip tracing knowledge base according to the object identifier if a new video clip request is detected, determining the historical video clip data and object preference data, and recommending clip elements based on the historical video clip data and object preference data, wherein the historical video clip data includes: historical structured video clip data.

[0170] The description of each module above refers to the corresponding description in the method embodiment and will not be repeated here.

[0171] According to the video editing tracing knowledge base construction device provided by the embodiment of the present application, by recording the relevant video editing data in the editing process into the video editing tracing knowledge base in real time, it is convenient to comprehensively and efficiently trace the editing history, which can effectively save users' search time and eliminate the need for users to spend a lot of time searching, for example, searching for materials, adjusting effects, etc., thereby improving the quality and efficiency of video production. In addition, it can also locate and repair problems, and ensure the security and compliance of data use.

[0172] An embodiment of the present application provides a non-volatile computer storage medium, which stores at least one executable instruction or computer program, which can enable a processor to perform operations corresponding to the video clip tracing knowledge base construction method in any of the above method embodiments.

[0173] An embodiment of the present application provides a computer program product, which includes at least one executable instruction or computer program, which can enable a processor to perform operations corresponding to the video clip tracing knowledge base construction method in any of the above method embodiments.

[0174] Figure 4 A schematic structural diagram of an embodiment of a computing device of the present application is shown. The specific embodiment of the present application does not limit the specific implementation of the computing device.

[0175] like Figure 4 As shown, the computing device may include: a processor 402 , a communications interface 404 , a memory 406 , and a communication bus 408 .

[0176] Processor 402, communication interface 404, and memory 406 communicate with each other via communication bus 408. Communication interface 404 is used to communicate with other devices, such as client devices or other server network elements. Processor 402 is used to execute program 410, which may specifically perform the steps described in the embodiment of the method for constructing a video clip tracing knowledge base for a computing device.

[0177] Specifically, the program 410 may include program codes, which include computer operation instructions.

[0178] Processor 402 may be a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application. The one or more processors included in the computing device may be processors of the same type, such as one or more CPUs, or may be processors of different types, such as one or more CPUs and one or more ASICs.

[0179] The memory 406 is used to store the program 410. The memory 406 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0180] Program 410 can be specifically configured to cause processor 402 to execute the video clip tracing knowledge base construction method described in any of the above-described method embodiments. The specific implementation of each step in program 410 can be found in the corresponding descriptions of the corresponding steps and units in the above-described video clip tracing knowledge base construction embodiments, and will not be repeated here. Those skilled in the art will clearly understand that, for ease and brevity of description, the specific operating processes of the devices and modules described above can refer to the corresponding process descriptions in the above-described method embodiments, and will not be repeated here.

[0181] The algorithm and display provided herein are not inherently relevant to any particular computer, virtual system or other device. Various general-purpose systems can also be used together with the teachings based on this. According to the above description, it is obvious that the structure required for constructing this type of system. In addition, the embodiments of the present application are not directed to any specific programming language yet. It should be understood that various programming languages can be utilized to realize the content of the embodiments of the present application described herein, and the description of the specific languages above is for the purpose of disclosing the best mode of implementation of the embodiments of the present application.

[0182] In the description provided herein, a large number of specific details are described. However, it is understood that the embodiments of the present application can be practiced without these specific details. In some instances, well-known methods, structures, and techniques are not shown in detail so as not to obscure the understanding of this description.

[0183] Similarly, it should be understood that, for the purpose of streamlining the present disclosure and facilitating the understanding of one or more of the various inventive aspects, in the foregoing description of the exemplary embodiments of the embodiments of the present application, the various features of the embodiments of the present application are sometimes grouped together into a single embodiment, figure, or description thereof. However, the disclosed method should not be construed as reflecting an intention that the claimed embodiments of the present application require more features than are expressly recited in each claim. Rather, as reflected in the following claims, the inventive aspects lie in less than all the features of the single foregoing disclosed embodiment. Thus, the claims following the detailed description are hereby expressly incorporated into the detailed description, with each claim standing on its own as a separate embodiment of the embodiments of the present application.

[0184] Those skilled in the art will appreciate that the modules in the devices in the embodiments can be adaptively changed and disposed in one or more devices different from the embodiments. The modules or units or components in the embodiments can be combined into one module or unit or component, and in addition, they can be divided into multiple sub-modules or sub-units or sub-components. Any combination can be adopted for all the features disclosed in this specification (including the accompanying claims, abstract, and drawings) and all the processes or units of any method or device so disclosed, except that at least some of such features and / or processes or units are mutually exclusive. Unless otherwise expressly stated, each feature disclosed in this specification (including the accompanying claims, abstract, and drawings) can be replaced by an alternative feature that provides the same, equivalent, or similar purpose.

[0185] In addition, those skilled in the art will be able to understand that, although some of the embodiments described herein include certain features included in other embodiments but not others, the combination of the features of different embodiments means that it is within the scope of the embodiments of the present application and forms different embodiments. For example, in the following claims, any one of the claimed embodiments can be used in any combination.

[0186] The various component embodiments of the embodiments of the present application can be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. It should be understood by those skilled in the art that a microprocessor or digital signal processor (DSP) can be used in practice to implement some or all of the functions of some or all of the components in the embodiments of the present application. The embodiments of the present application can also be implemented as a device or apparatus program (e.g., computer program and computer program product) for performing a part or all of the methods described herein. Such a program implementing the embodiments of the present application can be stored on a computer-readable medium, or can have the form of one or more signals. Such a signal can be downloaded from an Internet website, or provided on a carrier signal, or provided in any other form.

[0187] It should be noted that the above embodiments illustrate rather than limit the embodiments of the present application, and that a person skilled in the art may devise alternative embodiments without departing from the scope of the appended claims. In the claims, any reference symbols placed between brackets should not be construed as limiting the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The embodiments of the present application may be implemented by means of hardware comprising several different elements and by means of a suitably programmed computer. In a unit claim enumerating several means, several of these means may be embodied by the same item of hardware. The use of the words first, second, and third, etc. does not indicate any order. These words may be interpreted as names.

Claims

1. A method for constructing a video editing traceability knowledge base, comprising: Obtaining video editing data in real time; Performing data structuring processing on the video editing data to obtain structured video editing data; Performing data integration and analysis processing on the structured video editing data to generate knowledge graph data, wherein the knowledge graph data includes: at least two nodes, the association relationships between the at least two nodes, and node attributes; Storing the structured video editing data and the knowledge graph data in a video editing traceability knowledge base according to a preset data structure.

2. The method according to claim 1, wherein, The video editing traceability knowledge base includes: a relational database and a graph database; The storing the structured video editing data and the knowledge graph data in a video editing traceability knowledge base according to a preset data structure further includes: Storing the structured video editing data in the relational database according to a first preset data structure; Storing the knowledge graph data in the graph database according to a second preset data structure.

3. The method according to claim 1 or 2, wherein, The video editing data includes: video material data, editing behavior data, and editing draft data; The nodes include: video material nodes, editing behavior nodes, and draft version nodes; The association relationships include: the relationship between video materials and editing behaviors and the relationship between editing behaviors and draft versions.

4. The method according to any one of claims 1 to 3, wherein The method further includes: receiving a first traceability request, wherein the first traceability request includes: first traceability information; Analyzing the first traceability information, extracting first traceability keywords, and performing intention recognition processing on the first traceability information to determine a traceability intention; Retrieving the video editing traceability knowledge base according to the first traceability keywords and the traceability intention to obtain a first video editing traceability result; Displaying the first video editing traceability result in a preset form in a user interface; and / or Performing natural language conversion processing on the first video editing traceability result and displaying the conversion result in the user interface.

5. The method according to claim 4, wherein, The traceability intention includes: editing behavior traceability, draft traceability, and video material traceability.

6. The method according to claim 4 or 5, wherein, Displaying the first video editing traceability result in a preset form in a user interface; And / or, after performing natural language conversion processing on the first video editing traceability result and displaying the conversion result in the user interface, the method further includes: Continuing to obtain at least one second traceability request, wherein the second traceability request includes: second traceability information; For any second traceability information, analyzing the second traceability information, extracting second traceability keywords, and retrieving the video editing traceability knowledge base according to the second traceability keywords to obtain a second video editing traceability result; Performing combination processing on the first video editing traceability result and the second video editing traceability result and displaying the combination processing result in a preset form in the user interface; and / or Performing natural language conversion processing on the first video editing traceability result and at least one second video editing traceability result and displaying the conversion result in the user interface.

7. The method according to any one of claims 1-6, wherein The method further includes: receiving a draft comparison request, wherein the draft comparison request includes a draft identifier to be compared; According to the draft identifier to be compared, the corresponding video drafts in the video clip tracing knowledge base are analyzed and compared to generate draft differences; The draft differences are presented in a preset form in a user interface.

8. The method according to any one of claims 1 to 7, wherein: The method further includes: if a new video editing request is detected, searching a video editing tracing knowledge base according to an object identifier, determining historical video editing data and object preference data, and recommending editing elements based on the historical video editing data and object preference data, wherein the historical video editing data includes historical structured video editing data.

9. A video clip tracing knowledge base construction device, comprising: A data acquisition module, suitable for acquiring video clip data in real time; a data structuring processing module, adapted to perform data structuring processing on the video clipping data to obtain structured video clipping data; a data integration analysis processing module, adapted to perform data integration analysis on the structured video clip data to generate knowledge graph data, wherein the knowledge graph data includes: at least two nodes, an association relationship between the at least two nodes, and node attributes; The video clip tracing knowledge base is suitable for storing structured video clip data and the knowledge graph data according to a preset data structure.

10. A computing device, comprising: A processor, a memory, a communication interface, and a communication bus, wherein the processor, the memory, and the communication interface communicate with each other via the communication bus; The memory is used to store at least one executable instruction, and the executable instruction enables the processor to perform operations corresponding to the method for constructing a video clip tracing knowledge base according to any one of claims 1 to 8.

11. A computer storage medium, wherein at least one executable instruction is stored in the storage medium, wherein the executable instruction enables a processor to execute operations corresponding to the method for constructing a video clip tracing knowledge base according to any one of claims 1 to 8.

12. A computer program product comprising at least one executable instruction, wherein the executable instruction enables a processor to execute operations corresponding to the method for constructing a video clip tracing knowledge base according to any one of claims 1 to 8.

Images