Digital culture narrative content generation methodology and system based on user co-creation and semantic real-time interaction

Abstract

The invention discloses a digital culture narrative content generation method and system based on user co-creation and semantic real-time interaction, and belongs to the technical field of man-machine interaction and intelligent media. The objective of the invention is to solve the integration problem that in existing cultural heritage digital display, users participate in one direction, the creation threshold is high, and intelligent real-time interaction with collection resources cannot be carried out. The method comprises the steps of obtaining a user original text and performing semantic analysis to generate a unified semantic vector; synchronously retrieving a cultural heritage digital resource library and a dynamic special effect library according to the vector to obtain a matched material set; fusing the material and the user image to generate an initial narrative medium; responding to a user real-time editing instruction, re-retrieving and updating the material based on the maintained semantic context, and generating an update preview in real time through local re-rendering; and finally outputting a personalized narrative work and a digital commemorative carrier. Through a closed-loop semantic interaction architecture, user creative input, AI semantic understanding, intelligent resource scheduling and real-time rendering output are organically integrated into a collaborative system, and technical normal form transformation from one-way viewing to two-way real-time co-creation is achieved.

Description

Technical Field

[0001] This invention belongs to the field of human-computer interaction and intelligent media generation technology, specifically involving a method, system and storage medium for collaborative creation of digital content based on user natural language input, artificial intelligence semantic understanding and real-time interactive editing. Its technical solution is particularly suitable for the personalized generation and dissemination of digital narrative content of cultural heritage. Background Technology

[0002] In the field of digital dissemination of cultural heritage, existing technologies mainly focus on two directions. The first direction emphasizes high-fidelity digital recording and one-way display of the cultural heritage itself, such as constructing digital archives or virtual exhibition halls through technologies like 3D laser scanning and panoramic photography. While such technologies enable remote access and immersive observation, their core paradigm remains "institutional display, public viewing," with users in a passive receiving state, lacking channels for deep participation and personal emotional expression.

[0003] The second approach attempts to introduce user interaction, but these are mostly limited to simple operations within preset paths, such as setting up artifact puzzles, quizzes, or photo filters based on fixed templates in mobile applications. While such solutions enhance the fun, they do not give users true creative autonomy, resulting in highly homogenized content that fails to form personalized digital assets that carry personal narratives.

[0004] In recent years, generative artificial intelligence technology has made progress in image and video synthesis. However, current applications of this technology in the field of cultural heritage are mostly style transfer or static generation. No technological solution has yet been found that can systematically integrate the following three elements: free-form creative input from users (such as short poems), a vast digital resource library of cultural heritage, and an interactive generation engine that supports real-time semantic-level intervention by users in the generation process. A common problem with existing technologies is that the technical means are fragmented, failing to form a complete, closed-loop, real-time interactive collaborative creation system from creative input to personalized finished product output. Summary of the Invention

[0005] Purpose of the Invention: To address the issues of technological fragmentation and lack of interaction in the aforementioned background technologies, this invention proposes a method and system for generating digital cultural narrative content based on user co-creation. The core of this method lies in organically integrating user creativity, AI analysis, resource scheduling, and media synthesis through a real-time interactive closed loop that maintains semantic context consistency.

[0006] Technical Solution: This invention aims to solve the problems of technological fragmentation and lack of interaction in the above-mentioned background technology, and proposes a digital cultural narrative content generation scheme based on user co-creation and real-time semantic interaction. Its core lies in the organic integration of user creativity, artificial intelligence analysis, intelligent resource scheduling and media synthesis through a real-time interactive closed loop that maintains semantic context consistency.

[0007] To achieve the above objectives, the present invention adopts the following technical solution: In a first aspect, the present invention provides a method for generating digital cultural narrative content based on user co-creation. This method is implemented through a real-time interactive closed loop that maintains semantic context consistency, and includes the following steps: S1. Obtain original text content and personal image data entered by the user through the interactive interface.

[0008] S2. Semantic parsing and vectorization of the original text content: Using a natural language processing model finely tuned to a corpus in the field of cultural heritage, the text is parsed and its emotional dimension features and cultural heritage image entities are extracted. The features and entities are mapped to the same high-dimensional vector space to generate a unified semantic vector.

[0009] S3. Intelligent resource matching: Using the unified semantic vector as the query condition, perform semantic matching based on cosine similarity on the pre-constructed cultural heritage digital resource library and dynamic special effects material library simultaneously to obtain a set of matching digital materials.

[0010] S4. Initial Content Composition: Based on the preset narrative time and space template, the matched digital material set, personal image data and related dynamic effects are layered and timelined to generate the initial narrative media file.

[0011] S5. Real-time semantic interactive editing: The interactive interface provides a preview and responds to user selection and replacement commands for any point; based on the semantic context maintained by the unified semantic vector, the system initiates a new round of intelligent resource matching in real time, obtains alternative materials, and drives the content synthesis engine to perform partial re-rendering of the initial narrative media file to generate an updated preview.

[0012] S6. Final Output: In response to the user's confirmation command, output the final personalized narrative media file and its associated digital memorial carrier.

[0013] Preferably, the natural language processing model in step S2 is a pre-trained model based on the Transformer architecture, which outputs sentiment dimension features through a classifier and extracts cultural heritage image entities through a named entity recognition module.

[0014] Preferably, the semantic matching in step S3 is achieved by calculating the cosine similarity between the unified semantic vector and the material tag vector, and selecting materials with a similarity higher than a preset threshold.

[0015] Preferably, the local re-rendering in step S5 is performed without interrupting the media stream playback, only replacing and recompositing the material layer targeted by the point-and-replace command.

[0016] Secondly, the present invention provides a digital cultural narrative content generation system for implementing the above-described method, comprising: The user interface module is used to obtain user input and editing instructions and display a preview. A semantic management engine is used to perform semantic parsing and vectorization, and to generate and maintain the unified semantic vector and semantic context. The resource scheduling engine connects to the cultural heritage digital resource library and the dynamic special effects material library to perform intelligent matching and update retrieval based on unified semantic vectors; A media compositing engine for performing spatiotemporal blending and real-time local re-rendering; The real-time interactive control module is used to capture user commands, coordinate the semantic management engine and resource scheduling engine to perform real-time semantic interactive editing, and control the media compositing engine; The output module is used to generate and output the final file.

[0017] Thirdly, the present invention provides an electronic device including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the program, implements the method as described in the first aspect.

[0018] Fourthly, the present invention provides a computer-readable storage medium having a computer program stored thereon that, when executed by a processor, implements the method described in the first aspect.

[0019] Beneficial effects: Compared with the prior art, the present invention has the following beneficial effects: 1. Real-time interactive editing with semantic consistency is achieved, overcoming the technical shortcomings of traditional non-linear editing, such as low efficiency and semantic breaks between different parts of the text in matching professional cultural heritage materials.

[0020] 2. By using a unified semantic vector to drive synchronous retrieval across multiple resource libraries, the system's response efficiency and the accuracy of material association have been improved.

[0021] 3. It has formed a closed-loop workflow of "creation-preview-editing", integrating multi-step professional processes into a single and coherent operation, significantly reducing the complexity of user operations, and promoting the paradigm shift of users from cultural receivers to cultural co-creators.

[0022] Advantages: Compared with the prior art, the present invention has the following advantages: 1. It realizes real-time interactive editing for maintaining semantic consistency, overcoming the technical defects of low efficiency and semantic discontinuity before and after in traditional non-linear editing software in the matching of professional cultural heritage materials; 2. By driving synchronous retrieval of multiple resource libraries with unified semantic vectors, it improves the system response efficiency and the accuracy of material association; 3. It forms a closed-loop workflow of "creation - preview - editing", reducing the complexity of user operations and integrating multi-step professional processes into a single coherent operation. Description of the Drawings

[0023] Figure 1 It is a flowchart of the method for generating digital cultural narrative content provided by an embodiment of the present invention.

[0024] Figure 2 It is a schematic diagram of the principle of semantic vector matching and real-time recommendation in an embodiment of the present invention.

[0025] Figure 3 It is a schematic diagram of the real-time interactive editing interface and the internal response data stream of the system in an embodiment of the present invention. Detailed Embodiments

[0026] To make the objectives, technical solutions and advantages of the present invention clearer, the following will describe the embodiments of the present invention in detail in conjunction with the drawings. The following embodiments are only used to explain the present invention and do not limit the protection scope of the present invention.

[0027] Embodiment: Take the example of a user visiting a jade exhibition and creating a poem. The user inputs the poem "Jade pendant condenses the moonlight". The system obtains this text through the user interface module (S1). The semantic management engine starts the fine-tuned BERT model for in-depth parsing, outputs the emotion "serenity" and the image "ancient jade bi", and maps and generates a high-dimensional semantic vector V1 (S2). The resource scheduling engine uses V1 as the query vector to synchronously retrieve the jade digital resource library and the Chinese-style dynamic special effect library, and matches the "Warring States valley pattern jade bi" model and the "hazy halo" special effect with the highest similarity by calculating the cosine similarity between V1 and the feature vectors of the materials in the library (S3). The media synthesis engine fuses the user's personal photo, the matched jade bi model and the halo special effect according to the template to generate an initial short film (S4).

[0028] When a user selects the jade disc model in the preview with the intention of replacing it, the real-time interactive control module captures this instruction. The semantic management engine combines the object label with V1 to generate a refined query vector V1′. The resource scheduling engine recalculates the cosine similarity based on V1′ and retrieves a list of alternative materials in real time, such as "Qing Dynasty white jade dragon disc" (S5). After the user makes a selection, the media compositing engine performs millisecond-level local re-rendering and replacement of the jade disc layer without interrupting the preview. Throughout the process, the "tranquil jade realm" defined by the initial semantic vector V1 is continuously maintained as context to ensure that all replacements do not deviate from the original poetic meaning. After the user confirms, the output module generates the final short film and digital commemorative card (S6).

[0029] Those skilled in the art will understand that the model architecture, resource library composition, and specific parameters in the above embodiments can be adjusted according to actual needs, and these modifications and variations all fall within the protection scope defined by the claims of this invention.

Claims

1. A method for generating digital cultural narrative content based on user co-creation, characterized in that, This is achieved through a real-time interactive closed loop that maintains semantic context consistency, including: • Obtain original text content input by the user; • Semantic analysis is performed on the original text content to generate a unified semantic vector containing both emotional and cultural heritage imagery dimensions, and this unified semantic vector is used as the semantic anchor point for the entire creative process; • Using the unified semantic vector as the query condition, the cultural heritage digital resource library and dynamic special effects library are searched simultaneously to obtain an initial set of semantically matched digital materials; • The initial set of digital materials and the user-provided personal video data are spatiotemporally fused according to a preset narrative template to generate an initial narrative media file; • In response to a user's selection and replacement instruction for any digital material in the initial narrative media file, based on the semantic tags of the selected material and using the unified semantic vector as the retrieval baseline, alternative materials with semantic similarity higher than a preset threshold are retrieved from the cultural heritage digital resource library, and the digital material set is updated. Based on the updated set of digital materials, without interrupting the playback of the narrative media file stream, the layers corresponding to the replaced materials are locally re-rendered in real time to generate updated narrative media files for preview. • Output the final narrative media file and its associated digital memorial carrier.

2. The method according to claim 1, characterized in that, The semantic parsing of the original text content is performed by using a pre-trained language model fine-tuned for the cultural heritage field, which simultaneously outputs sentiment classification labels and cultural heritage image entity recognition results, and maps the two to the same vector space to form the unified semantic vector.

3. The method according to claim 1, characterized in that, In the synchronous retrieval, the cosine similarity between the unified semantic vector and the material tag vectors in the cultural heritage digital resource library and the dynamic special effects library is calculated, and materials with similarity higher than a preset threshold are used as the set of matching digital materials.

4. The method according to claim 1, characterized in that, The response to the user's real-time selection and replacement instruction further includes: the system performing semantic similarity retrieval in the cultural heritage digital resource database based on the semantic tags of the selected material, with the unified semantic vector as the context constraint, and presenting a list of alternative materials for the user to choose from, sorted by similarity.

5. The method according to claim 1, characterized in that, The real-time re-rendering involves updating the layers corresponding to the replaced material at the local pixel level while maintaining continuous playback of the narrative media file, and then compositing them with the unreplaced layers in real time to achieve a seamless preview of the editing results.

6. A digital cultural narrative content generation system, used to implement the method of any one of claims 1-5, characterized in that, include: • The user interface module is used to obtain original text content, personal image data, and real-time selection and replacement instructions for digital materials in narrative media files input by users; • Semantic management engine, used to perform semantic parsing on the original text content, generate and persist a unified semantic vector containing emotional and cultural heritage imagery dimensions, and maintain the semantic context throughout the interaction process; • A resource scheduling engine, connected to the cultural heritage digital resource library and the dynamic effects library, is used to perform initial synchronous retrieval based on the unified semantic vector and to perform updated retrieval based on semantic similarity according to real-time editing instructions; • Media compositing engine, used to spatiotemporally merge digital material collections with personal video data according to preset narrative templates, and perform local real-time re-rendering without interrupting playback; • The real-time interactive control module is used to capture the user's click-to-replace command, coordinate the semantic management engine and the resource scheduling engine to complete the semantic-driven material update, and trigger the media compositing engine to re-render; • Output module, used to generate and output the final narrative media file and its associated digital memorial carrier.

7. An electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the method as described in any one of claims 1-5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the method as described in any one of claims 1-5.

Images