Non-perpetual culture AR (Augmented Reality) interactive display system fusing virtual digital human and real scene

Abstract

The invention discloses a non-perpetual culture AR interactive display system fusing a virtual digital human and a real scene, and relates to the technical field of computer graphics virtual enhancement, and the system comprises a real scene collection module which collects scene data and processes the scene data to generate a three-dimensional model and a texture map; the non-perpetual digital resource library adopts distributed storage to classify and store standardized non-perpetual resources and construct indexes; the virtual digital human construction module constructs a model and an action library, and carries a language processing engine and a knowledge graph; the AR fusion rendering module realizes virtual and real registration, and a fusion effect is guaranteed through an illumination adaptation algorithm; the interaction module supports multi-mode interaction; the scene adaptation optimization module dynamically adjusts parameters to adapt to different environments; and the data storage management module guarantees the data security through encryption, authority control and backup mechanisms. According to the method, virtuality and reality fusion and natural coordination are achieved, the interaction mode is rich and immersive, non-perpetual content transmission is accurate and authoritative, and non-perpetual culture live inheritance and wide propagation are assisted.

Description

Technical Field

[0001] This invention relates to the field of computer graphics virtual augmentation technology, and in particular to an AR interactive display system for intangible cultural heritage that integrates virtual digital humans with real-world scenes. Background Technology

[0002] Intangible cultural heritage (ICH), as an important carrier of national culture, encompasses various categories such as traditional crafts, performing arts, and folk festivals, carrying unique historical value and cultural connotations. However, the current inheritance of ICH faces many difficulties: most ICH projects are located in remote areas, and due to geographical limitations, offline display scenarios are limited. Traditional display methods such as museums and ICH halls mainly rely on static exhibits and textual explanations, resulting in a monotonous presentation. Visitors can only view from a distance and cannot experience the details of the craftsmanship or engage in interactive experiences up close. Some ICH skills rely on oral transmission, and with the aging of inheritors, many skills face the risk of being lost. Traditional inheritance models are inefficient and difficult to disseminate on a large scale. Young people lack sufficient channels to access and understand ICH, leading to a gradual shrinking audience for ICH culture.

[0003] To overcome the limitations of dissemination, some technologies have attempted to apply AR technology to the display of intangible cultural heritage. However, existing solutions still have significant shortcomings: the integration of virtual and real elements is poor. Most systems can only achieve simple overlay of virtual images onto real scenes, lacking adaptation and optimization for lighting, proportions, movements, and environment. The virtual objects and real scenes are strongly disconnected, resulting in a stiff visual experience. The design of virtual digital humans is crude, mostly using generic 3D models that do not incorporate the real image and skill characteristics of intangible cultural heritage inheritors. Their facial expressions are stiff, their movements lack professionalism, and they cannot accurately reproduce the operational details of intangible cultural heritage skills. Furthermore, they lack intangible cultural heritage knowledge, making it difficult to conduct in-depth Q&A and explanations. The interaction methods are limited to basic operations such as clicking and swiping on touchscreens, lacking multimodal interaction methods such as gesture recognition and dialect voice interaction. This results in insufficient user immersion, and the fixed interactive storylines cannot generate personalized content based on user preferences, making it difficult to stimulate user participation.

[0004] Furthermore, there are also problems with the management and system adaptability of intangible cultural heritage digital resources: Intangible cultural heritage resources are stored in a scattered manner, lacking a unified classification and indexing system, making retrieval and retrieval inconvenient, and the resource update mechanism is imperfect, failing to incorporate the latest research results and updated skills of inheritors in a timely manner; the system's adaptability is weak, easily experiencing lag and rendering distortion under different lighting environments, different performance devices, or low-bandwidth networks, affecting user experience; data security and traceability are insufficient, user interaction records, resource call logs, and other data are easily lost or tampered with, making it difficult to guarantee the authority of intangible cultural heritage resources, and there is a lack of a quantitative evaluation mechanism for user experience, making it impossible to optimize system performance in a targeted manner. These problems prevent existing AR display systems from fully leveraging their technological advantages and meeting the needs of the living transmission, widespread dissemination, and in-depth education of intangible cultural heritage. There is an urgent need for a highly integrated, highly interactive, and adaptable AR interactive display solution for intangible cultural heritage. Summary of the Invention

[0005] The present invention proposes an AR interactive display system for intangible cultural heritage that integrates virtual digital humans with real-world scenes, in order to solve the problems mentioned in the prior art.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: an AR interactive display system for intangible cultural heritage that integrates virtual digital humans with real-world scenes, comprising the following modules: Real-world scene acquisition module: Equipped with a 3D laser scanner, camera and GPS positioning module, it acquires the 3D coordinates, topography, exhibit parameters, scene images and dynamic images of the intangible cultural heritage display scene; it performs gamma correction, noise reduction filtering and image stitching processing on the acquired data to generate a 1:1 scale 3D point cloud model and texture map; The Intangible Cultural Heritage Digital Resource Database adopts a distributed storage architecture to classify and store digital resources of intangible cultural heritage in the categories of traditional skills, performances, and folk festivals; the resources adopt a standardized format and are indexed by keywords, intangible cultural heritage categories, and regional attributes; Virtual Digital Human Construction Module: Collects facial features, body posture and motion data, uses next-generation 3D modeling technology to construct a digital human model, binds 52 muscle units to the face, records skill demonstrations and interactive postures through motion capture technology, builds a dedicated motion library, and is equipped with a natural language processing engine and intangible cultural heritage knowledge graph; AR fusion rendering module: integrates SLAM technology to realize digital human registration in virtual and real scenes, and coordinates virtual and real lighting and shadows through lighting adaptation algorithms; Interactive module: Supports gesture, voice, and touch interaction; voice recognition covers Mandarin and 20+ dialects; touch is compatible with mobile terminals and supports click and swipe operations; Scene adaptation and optimization module: Real-time monitoring of ambient light, device performance and network bandwidth; night mode is enabled when light is ≤500 lux; model accuracy is reduced when device performance is insufficient; incremental transmission is used when bandwidth is ≤10Mbps. Data storage management module: Blockchain stores key data; distributed database stores raw data of scenario collection, models and resources; AES-256 encryption is used to set access permissions and establish a backup mechanism.

[0007] Furthermore, it also includes a virtual digital human and scene adaptation optimization unit, through... Calculate the virtual-real fusion adaptation degree, where F is the fusion adaptation degree value. For lighting adaptation weights, For lighting matching degree, To adapt the weights proportionally, For the sake of proportional harmony, To adapt weights to actions, For motor coordination, For environment adaptation weights, For environmental compatibility.

[0008] Furthermore, it also includes an interactive story generation module: an interactive story framework is built based on the core content of intangible cultural heritage, including the main storyline and side storylines. The system dynamically generates customized storylines based on user interaction behavior, selection preferences and dwell time. Interactive elements are set at story nodes, and the results of the interaction affect the direction of the storyline. Explanations of intangible cultural heritage knowledge points are embedded in the storyline.

[0009] Furthermore, it also includes an interactive response priority scheduling unit, through... Determine the order of interaction request responses, where Prioritize interactive responses. In response to timeliness weighting, To request the urgency level, As a weight for content importance, To enhance the value of intangible cultural heritage through interactive content, Weighting based on user demand To increase user activity As the system load weight, This represents the current system resource utilization rate; the higher the priority value, the faster the response.

[0010] Furthermore, the virtual digital human construction module also includes an intelligent learning unit: by collecting user interaction feedback data, the latest research results on intangible cultural heritage, and updated content on the skills of inheritors, deep learning algorithms are used to optimize the language expression, action demonstration, and knowledge reserves of the virtual digital human. The virtual digital human action library and knowledge graph are updated quarterly, and the model accuracy is iterated and upgraded once a year.

[0011] Furthermore, it also includes a user experience evaluation unit, through... Quantify user experience effects, among which The overall score for user experience To incorporate smoothness weights, The smoothness of the virtual-real integration is scored. Weighting for interactive immersion, Rate the level of interactive immersion. As a weight for content satisfaction, Rate the satisfaction with intangible cultural heritage content. Weighting of learning outcomes The system scores the learning outcomes of intangible cultural heritage knowledge and dynamically adjusts system parameters based on the evaluation results.

[0012] Furthermore, the AR fusion rendering module also integrates a virtual-real collision detection unit: it uses the AABB bounding box algorithm to construct collision objects between virtual digital humans, virtual props and real scenes. The collision detection frequency is consistent with the rendering frame rate. When the distance between the virtual object and the real scene physical exhibit is less than 5 centimeters, the collision response mechanism is automatically triggered to adjust the position or transparency of the virtual object.

[0013] Furthermore, the intangible cultural heritage digital resource database also includes a resource update unit: establishing a dynamic update mechanism for intangible cultural heritage resources, establishing data connection channels with intangible cultural heritage protection centers, inheritors and academic research institutions, obtaining the latest intangible cultural heritage skill videos, documents and inheritor stories in real time, reviewing and confirming resources before updates, and notifying users through the system push function after updates, guiding users to experience the new content, retaining historical version resources, and supporting users to view and compare.

[0014] Furthermore, the interactive module also supports multi-user collaborative interaction: multiple users enter the same AR interactive scene through network connection, the system assigns independent virtual identity identifiers, and distributed rendering technology is used during multi-user interaction to record the multi-user interaction process and generate interactive videos, which can be shared and saved by users to expand the scope of intangible cultural heritage dissemination.

[0015] Furthermore, it also includes a scene expansion module: it supports users to customize intangible cultural heritage AR display scenes, provides a scene editor tool, allows users to import custom scene models, add virtual props and set interaction rules, the scene editor has a built-in intangible cultural heritage element material library, supports drag-and-drop operation and parameter adjustment, and after the custom scene is generated, it is published after being reviewed by the system for other users to experience.

[0016] Compared with existing technologies, the beneficial effects of this invention are: The integration of virtual and real elements is natural and harmonious. The real scene acquisition module generates a 1:1 scale 3D point cloud model and high-definition texture map through high-precision scanning and image processing, laying the foundation for virtual-real integration. The AR fusion rendering module combines SLAM technology to achieve accurate registration, and the lighting adaptation algorithm dynamically adjusts the parameters of virtual objects. Combined with the quantitative calculation of the matching optimization unit, the virtual digital human is highly consistent with the real scene in terms of lighting, scale, and movement, with outstanding visual consistency, which completely changes the problem of the separation between virtual and real in traditional AR systems.

[0017] The interactive experience is rich and immersive. The system supports multimodal interaction methods such as gestures, voice, and touch, covering the operating habits of different user groups. Interactive response priority scheduling ensures rapid response to core needs and high smoothness. The interactive story generation module dynamically constructs customized storylines based on user behavior, combining main and side storylines and embedding explanations of intangible cultural heritage knowledge points to achieve a deep integration of entertainment and popular science, greatly enhancing user participation and immersion, and solving the pain points of traditional display interaction being monotonous and lacking personalization.

[0018] The delivery of intangible cultural heritage content is accurate and authoritative. The virtual digital human is built based on the real image of the inheritor. The high-precision model and exclusive motion library accurately reproduce the details of the skills. The intangible cultural heritage knowledge graph and intelligent learning unit can continuously update the knowledge reserves and skill demonstration content to ensure the accuracy and timeliness of the information. The intangible cultural heritage digital resource library adopts classified storage and multi-level indexing. The resource update mechanism and expert review process ensure the authority of the content. Users can quickly retrieve and call the resources they need, promoting the accurate inheritance of intangible cultural heritage skills.

[0019] The system boasts strong adaptability and stability. The scene adaptation and optimization module monitors environmental and device parameters in real time, dynamically adjusting rendering resolution, data transmission rate, and other parameters. It can operate stably under different lighting, device, and network conditions, avoiding issues such as lag and distortion. The data storage management adopts a hybrid architecture of blockchain and distributed database, achieving encrypted data storage, immutability, and traceability. A multi-level backup mechanism ensures data security, and multi-level access permissions differentiate to meet the operational needs of different users.

[0020] Furthermore, the system supports multi-user collaborative interaction and user-defined scenes, allowing multiple users to collaborate on skill simulation tasks and generate interactive videos for easy sharing and dissemination. The scene editor and intangible cultural heritage element library lower the barrier to creation, enabling users to publish custom scenes and participate in rating and commenting, fostering a healthy creative ecosystem and further expanding the reach of intangible cultural heritage. Overall, this system, through the deep integration of technology and culture, provides an innovative path for the inheritance of intangible cultural heritage, balancing cultural dissemination, education, and entertainment, thus helping intangible cultural heritage achieve living transmission and widespread dissemination in the digital age. Attached Figure Description

[0021] Figure 1This is a schematic block diagram of the AR interactive display system for intangible cultural heritage that integrates virtual digital humans with real-world scenes, as proposed in this invention. Figure 2 A bar chart comparing the adaptability of virtual-real fusion in different scenarios; Figure 3 A bar chart comparing the scores of the core dimensions of user experience; Figure 4 Line graph showing the growth in the scope of intangible cultural heritage dissemination. Detailed Implementation

[0022] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0023] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," and "counterclockwise," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0024] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, features defined with "first" and "second" may explicitly or implicitly include one or more of the stated features. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified. Furthermore, the terms "installed," "connected," and "linked" should be interpreted broadly; for example, they may refer to a fixed connection, a detachable connection, or an integral connection; they may refer to a mechanical connection or an electrical connection; they may refer to a direct connection or an indirect connection through an intermediate medium; and they may refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances. The invention will now be described in further detail with reference to the accompanying drawings.

[0025] Reference Figures 1 to 4 A non-material cultural heritage AR interactive display system that integrates virtual digital humans with real-world scenes includes the following modules: Real-world scene acquisition module: Equipped with a 3D laser scanner, a 4K high-definition camera, and a GPS positioning module, the 3D laser scanner has a scanning accuracy of 0.1 mm, acquiring the 3D spatial coordinates, topography, and geometric parameters of physical exhibits of the real intangible cultural heritage display scene. The 4K high-definition camera acquires scene images and dynamic videos at a frame rate of 60 frames per second. The GPS positioning module achieves accurate calibration of the scene location. The acquired data is processed by gamma correction, noise reduction filtering, and image stitching to generate a 1:1 scale real-world scene 3D point cloud model and high-definition texture map. The Intangible Cultural Heritage Digital Resource Database adopts a distributed storage architecture to classify and store digital resources of intangible cultural heritage. It covers traditional crafts such as embroidery, pottery, and wood carving, including tool models, production step breakdown animations, and process parameter documents; traditional performances such as opera, dance, and folk arts, including motion capture data, audio narration, and 3D models of costumes and props; and folk festivals such as temple fairs, paper-cutting, and shadow puppetry, including scene restoration models, historical background texts, and interactive story scripts. All resources adopt standardized data formats and construct a multi-level index system through keywords, intangible cultural heritage categories, and regional attributes to support fast retrieval and access. Virtual Digital Human Construction Module: Based on the real images of intangible cultural heritage inheritors, facial feature points, body posture data and movement habits are collected. Next-generation 3D modeling technology is used to construct a high-precision virtual digital human model. The face is bound with 52 muscle movement units to realize natural expression simulation. The motion capture technology is used to record the demonstration movements of intangible cultural heritage skills and interactive response postures to build a dedicated motion library containing 1,000+ movements. It is equipped with a natural language processing engine and intangible cultural heritage knowledge graph, and supports voice interaction, question and answer and skill explanation. The digital human model supports real-time rendering and real-time motion driving. AR Fusion Rendering Module: Integrates SLAM real-time localization and mapping technology to achieve accurate registration of real scenes and virtual digital humans, with registration errors controlled within 2 pixels. It adopts a lighting adaptation algorithm to dynamically adjust the lighting intensity, color temperature, and shadow effects of virtual digital humans and real scenes, achieving natural and harmonious virtual-real fusion. It supports multi-terminal adaptation, including mobile phones, tablets, AR glasses, and other devices. The rendering frame rate is maintained at more than 30 frames per second, and the resolution supports 4K output. It has a transparency adjustment function, which can adjust the display transparency of virtual digital humans and virtual props according to scene requirements. Interactive module: Supports multimodal interaction. The gesture recognition module captures user gestures through the camera with a recognition accuracy of over 95%, enabling functions such as virtual digital human interaction and simulation of intangible cultural heritage skills. The voice interaction module supports Mandarin Chinese and 20+ dialects with a response time of ≤0.5 seconds, enabling Q&A on intangible cultural heritage knowledge and issuing interactive commands. The touch interaction module is compatible with mobile terminal touchscreens, supporting operations such as clicking, swiping, and zooming, enabling virtual prop calls, scene switching, and interactive plot progression. Scene adaptation and optimization module: Real-time monitoring of ambient light intensity, device performance parameters and network bandwidth, dynamically adjusting rendering resolution, model detail level and data transmission rate. When the ambient light intensity is below 500 lux, night mode is automatically turned on to enhance the display brightness of virtual objects. When the device performance is insufficient, the number of model faces and texture accuracy are automatically reduced. When the network bandwidth is below 10 Mbps, incremental transmission is used to load core resources to achieve a stable interactive experience in different environments. Data storage management module: Adopting a hybrid storage architecture of blockchain + distributed database, the blockchain stores key data such as user interaction records and intangible cultural heritage resource call logs, ensuring data immutability and traceability. The distributed database stores scenario-collected data, virtual digital human models, and original data of the intangible cultural heritage resource library, supporting more than 1,000 data read and write requests per second. It is equipped with AES-256 encryption technology to encrypt transmitted and stored data, sets multi-level access permissions to distinguish the operation permissions of administrators, ordinary users, and visitors, and establishes a data backup mechanism to achieve daily incremental backup and weekly full backup.

[0026] This invention also includes a virtual digital human and scene adaptation optimization unit, through... Calculate the virtual-real fusion adaptation degree, where F is the fusion adaptation degree value, ranging from 0 to 100. The lighting adaptation weight is set to 0.3. This represents the illumination matching degree, with a value ranging from 0 to 100. To adjust the weight proportionally, a value of 0.25 is used. For proportional consistency, the value ranges from 0 to 100. The action is assigned a weight, with a value of 0.25. For motor coordination, the value ranges from 0 to 100. The environment adaptation weight is set to 0.2. The environmental fit score ranges from 0 to 100. This calculation dynamically adjusts the lighting parameters, model scale, movement range, and display position of the virtual digital human, allowing it to fully integrate into the real scene and improve visual consistency.

[0027] This invention also includes an interactive story generation module: an interactive story framework is constructed based on the core content of intangible cultural heritage, including a main storyline and side storylines. The main storyline revolves around the inheritance of intangible cultural heritage skills and historical stories, while the side storylines provide personalized interactive experiences. The system dynamically generates customized storylines based on user interaction behavior, selection preferences, and dwell time. Interactive elements such as intangible cultural heritage knowledge Q&A and skill simulation operations are set in the storyline nodes. The interaction results affect the direction of the storyline. Explanations of intangible cultural heritage knowledge points are embedded in the storyline to achieve the dual effect of "entertainment + popular science". The storyline data is synchronized to the data storage and management module in real time, supporting the saving of storyline progress and the ability to resume play from where it left off.

[0028] This invention also includes an interactive response priority scheduling unit, through... Determine the order of interaction request responses, where This sets the priority for interactive responses, with a value ranging from 1 to 10. As a weight for timeliness, a value of 0.3 is assigned. The urgency level of the request is indicated by a value from 1 to 10. This is the content importance weight, with a value of 0.25. The value of the interactive content intangible cultural heritage is rated from 1 to 10. The weight for user demand is set to 0.25. User activity level, with a value from 1 to 10. This represents the system load weight, with a value of 0.2. This represents the current system resource utilization rate, ranging from 1 to 10. Higher priority values ​​result in faster responses, ensuring quicker responses to core intangible cultural heritage interactive content and requests from high-demand users, thus optimizing the smoothness of the interactive experience.

[0029] In this invention, the virtual digital human construction module also includes an intelligent learning unit: by collecting user interaction feedback data, the latest research results on intangible cultural heritage, and the updated content of inheritors' skills, deep learning algorithms are used to optimize the virtual digital human's language expression, action demonstration, and knowledge reserves. The virtual digital human's action library and knowledge graph are updated every quarter, and the model accuracy is iterated and upgraded once a year. The virtual digital human continuously adapts to the current status of intangible cultural heritage inheritance and accurately conveys the latest intangible cultural heritage knowledge and skill details.

[0030] This invention also includes a user experience evaluation unit, through... Quantify user experience effects, among which This is a comprehensive score for user experience, ranging from 0 to 100. To incorporate the smoothness weight, a value of 0.3 is used. The smoothness of virtual-real integration is scored, with a value of 0-100. The weight for interactive immersion is set to 0.25. Rate the level of immersion in the interactive experience, on a scale of 0-100. The content satisfaction weight is set to 0.25. Rate the satisfaction with intangible cultural heritage content, with a value of 0-100. As a weight for learning effectiveness, a value of 0.2 is assigned. The system scores the learning effectiveness of intangible cultural heritage knowledge, ranging from 0 to 100. Based on the evaluation results, the system parameters are dynamically adjusted to optimize the rendering effect, interactive response speed, and content presentation.

[0031] In this invention, the AR fusion rendering module also integrates a virtual-real collision detection unit: it uses the AABB bounding box algorithm to construct collision objects between virtual digital humans, virtual props and real scenes. The collision detection frequency is consistent with the rendering frame rate. When the distance between the virtual object and the real scene physical exhibit is less than 5 cm, the collision response mechanism is automatically triggered to adjust the position or transparency of the virtual object to prevent the virtual object from penetrating the real entity. It generates haptic feedback signals to adapt to AR devices that support haptic feedback and enhance the realism of the interaction.

[0032] In this invention, the intangible cultural heritage digital resource database also includes a resource update unit: establishing a dynamic update mechanism for intangible cultural heritage resources, establishing data connection channels with intangible cultural heritage protection centers, inheritors and academic research institutions, obtaining the latest intangible cultural heritage skill videos, documents and inheritor stories in real time, reviewing and confirming the resources before updating to ensure the accuracy and authority of the content, notifying users through the system push function after the update, guiding users to experience the new content, retaining historical version resources, and supporting users to view and compare.

[0033] In this invention, the interactive module also supports multi-user collaborative interaction: multiple users can enter the same AR interactive scene through network connection, the system assigns independent virtual identity identifiers, supports real-time voice communication, action interaction and collaborative completion of intangible cultural heritage skill simulation tasks between users, distributed rendering technology is used during multi-user interaction to ensure the rendering smoothness of each user terminal, the multi-user interaction process is recorded to generate interactive videos, and users can share and save them to expand the scope of intangible cultural heritage dissemination.

[0034] This invention also includes a scene expansion module: it supports users to customize intangible cultural heritage AR display scenes, provides a scene editor tool, and allows users to import custom scene models, add virtual props, and set interaction rules. The scene editor has a built-in intangible cultural heritage element material library, including traditional patterns, classical architectural components, intangible cultural heritage tool models, etc. It supports drag-and-drop operation and parameter adjustment. After the custom scene is generated, it can be reviewed and published by the system for other users to experience. Users can rate and comment on the published scene, forming a positive and interactive intangible cultural heritage AR creation ecosystem.

[0035] The following two examples further illustrate the specific implementation of this system: Example 1: Application of AR Interactive Display System for Suzhou Embroidery Techniques in Offline Intangible Cultural Heritage Museums This example focuses on a Suzhou embroidery technique demonstration area in a provincial intangible cultural heritage museum. The area covers 80 square meters and displays over 30 items, including antique Suzhou embroidery pieces from the Ming and Qing dynasties, modern masterpieces, and traditional tools such as needles, frames, and threads. Traditional displays primarily rely on static exhibits and textual explanations, making it difficult for visitors to directly experience the details of Suzhou embroidery stitches and the production process. There is an urgent need to utilize AR technology to integrate virtual digital figures with real-world scenes, creating an immersive and interactive experience to facilitate the living transmission of Suzhou embroidery techniques.

[0036] I. System Deployment and Module Refinement The real-scene acquisition module deploys a 3D laser scanner, a 4K high-definition camera, and a GPS positioning module within the dedicated area. The 3D laser scanner achieves a scanning accuracy of 0.1 mm, capturing the 3D spatial coordinates, terrain features, and geometric parameters of the physical exhibits within the authentic intangible cultural heritage display scene. The 4K high-definition camera captures scene images and dynamic videos at a frame rate of 60 frames per second, and the GPS positioning module accurately calibrates the scene's location. The acquired data undergoes gamma correction to adjust brightness uniformity, median filtering to reduce image noise, and feature point matching to stitch together multi-view images. Ultimately, a 1:1 scale 3D point cloud model and high-definition texture map of the real scene are generated, ensuring high scene fidelity when the virtual digital human is integrated.

[0037] The Intangible Cultural Heritage Digital Resource Library utilizes a distributed storage architecture to build a dedicated digital resource library for Suzhou embroidery. It categorizes resources into three main categories: Traditional Techniques, including 3D models of tools and step-by-step animations of 20 core Suzhou embroidery stitches such as flat stitch, seed stitch, and gold thread stitch, with each animation lasting 5-10 seconds, and process parameter documents; Traditional Clothing, including 3D models of Ming and Qing dynasty Suzhou embroidery clothing and modern innovative clothing, embroidery pattern breakdown diagrams, and fabric material parameters; and Historical Culture, including texts on the origin and development of Suzhou embroidery, biographies of famous inheritors, restoration cases of ancient embroidery pieces, and interactive story scripts such as "Embroidery Maiden's Apprenticeship" and "Court Embroidery." All resources are standardized in formats such as FBX, MP4, and TXT, and a three-level index system is constructed using "stitch name + dynasty + purpose," supporting quick user keyword searches with a search response time of ≤0.3 seconds.

[0038] The virtual digital human module is modeled after the real image of a national-level Suzhou embroidery inheritor. It uses 3D scanning to collect over 5000 facial feature points, body posture data, and embroidery movement habits. Employing next-generation 3D modeling technology, it constructs a high-precision virtual digital human model with over 1 million faces and 52 muscle motion units bound to the face, capable of simulating natural expressions such as smiling and concentration. Motion capture technology records the inheritor's embroidery movements, including threading, needlework, and finishing, down to the angle of finger joint force, creating a dedicated motion library of over 1200 actions. It features a natural language processing engine and a Suzhou embroidery knowledge graph, covering over 1000 knowledge points including needlework techniques, thread selection, and historical anecdotes. Supporting interaction in both Mandarin and Wu dialects, it can accurately answer various user questions about Suzhou embroidery techniques. The digital human model supports real-time rendering and real-time motion-driven operation, following interactive commands without delay.

[0039] The AR fusion rendering module integrates SLAM real-time localization and mapping technology to achieve accurate registration between the real exhibition hall and the virtual digital human, with a registration error controlled within 2 pixels. It uses a lighting adaptation algorithm to dynamically adjust the virtual digital human parameters. When the natural lighting intensity of the exhibition hall changes, the skin reflections and clothing shadows of the virtual digital human adapt synchronously to ensure consistency between virtual and real lighting. It supports adaptation to three types of terminals: mobile phones, tablets, and AR glasses, maintaining a rendering frame rate of 35 frames per second, outputting 4K resolution, and featuring 0-100% transparency adjustment, allowing adjustment of the virtual digital human's display state according to the exhibit display requirements. The virtual digital human and scene adaptation optimization unit uses a formula... Calculate the virtual-real fusion adaptation degree, where The fit value is set to 0-100. The lighting adaptation weight is set to 0.3. This represents the illumination matching degree, with a value ranging from 0 to 100. To adjust the weight proportionally, a value of 0.25 is used. For proportional consistency, the value ranges from 0 to 100. The action is assigned a weight, with a value of 0.25. For motor coordination, the value ranges from 0 to 100. The environment adaptation weight is set to 0.2. This represents environmental fit, with a value ranging from 0 to 100. It was detected during a certain interaction. Substituting into the formula, we get Based on the calculation results, the system automatically fine-tunes the virtual digital human's head orientation and movement range, improving the fit to 92, making the virtual digital human visually blend more naturally with the exhibition hall environment.

[0040] The interactive module supports multimodal interaction: the gesture recognition module captures user gestures via the terminal camera with a recognition accuracy of 96%; when a user mimics the "flat stitch" motion, the virtual avatar simultaneously demonstrates the stitch. The voice interaction module supports Mandarin and Wu dialect recognition with a response time of 0.4 seconds; when a user asks "How to start a seed stitch?", the virtual avatar provides an immediate explanation and demonstration. The touch interaction module is compatible with touchscreens; users can switch between different stitch demonstrations by clicking the "stitch selection" button on the terminal interface, and zoom in / out on virtual tools by swiping the screen. The interactive response priority scheduling unit uses formulas... Determine the order of interaction request responses, where This sets the priority for interactive responses, with a value ranging from 1 to 10. As a weight for timeliness, a value of 0.3 is assigned. The urgency level of the request is indicated by a value from 1 to 10. This is the content importance weight, with a value of 0.25. The value of the interactive content intangible cultural heritage is rated from 1 to 10. The weight for user demand is set to 0.25. User activity level, with a value from 1 to 10. This represents the system load weight, with a value of 0.2. This represents the current system resource utilization rate, ranging from 1 to 10. A user simultaneously initiates a "flat stitch demonstration" request and a "Suzhou embroidery history" query. Substituting into the formula, we get This priority value is higher than historical query requests, and the system prioritizes responding to flat needle demonstrations to ensure the rapid implementation of core skills interaction.

[0041] The scene adaptation optimization and data storage management modules monitor the ambient light intensity, user terminal performance, and network bandwidth in real time: when the ambient light intensity is below 500 lux, a night mode is automatically activated to enhance the display brightness of virtual digital humans and props; when insufficient user terminal performance is detected, the polygon count of the virtual digital human model is automatically reduced from 1 million+ to 800,000, and the texture precision is reduced from 4K to 2K; when the network bandwidth is below 10Mbps, an incremental transmission method is used to load core actions and textures first, and then supplement detailed resources to ensure smooth interaction. The data storage management module adopts a hybrid architecture of blockchain and distributed database. The blockchain stores key data such as user interaction records and needle call logs, ensuring data immutability and traceability; the distributed database stores scene acquisition data, virtual digital human models, and raw data from the resource library, supporting 1200 data read / write requests per second. AES-256 encryption technology is used to encrypt transmitted and stored data, with three levels of access permissions: administrators can upload and update resources and modify system parameters; ordinary users can save interaction records and share experience videos; and visitors can only browse basic interactive content. Establish a data backup mechanism, perform incremental backups at 2 AM daily, and full backups every Sunday to prevent data loss.

[0042] Other functional modules' user experience evaluation unit uses formulas. Quantify user experience effects, among which This is a comprehensive score for user experience, ranging from 0 to 100. To incorporate the smoothness weight, a value of 0.3 is used. The smoothness of virtual-real integration is scored, with a value of 0-100. The weight for interactive immersion is set to 0.25. Rate the level of immersion in the interactive experience, on a scale of 0-100. The content satisfaction weight is set to 0.25. Rate the satisfaction with intangible cultural heritage content, with a value of 0-100. As a weight for learning effectiveness, a value of 0.2 is assigned. The learning effectiveness of intangible cultural heritage knowledge is scored, ranging from 0 to 100. Statistics are compiled after a particular operation. =90、 =88、 =92, L=85, substituting into the formula, we get... =0.3×90+0.25×88+0.25×92+0.2×85=27+22+23+17=89. Based on the evaluation results, the system fine-tunes the lighting adaptation algorithm parameters and interactive response thresholds to further optimize the experience. The resource update unit establishes a data connection channel with the Suzhou Embroidery Intangible Cultural Heritage Protection Center to obtain the latest skill videos and new embroidery data of inheritors in real time. After expert review, these are updated to the resource library, and historical versions are retained for users to view and compare. The scene expansion module provides a scene editor, allowing users to import custom exhibition hall layouts, add Suzhou embroidery patterns, tools, and other materials, set interactive rules, and publish the generated custom scenes for other users to experience after review.

[0043] II. System Implementation Process Scene acquisition and resource construction: Complete the 3D scanning and data processing of the Suzhou embroidery exhibition hall, and construct a 3D model of the real scene; organize digital resources of Suzhou embroidery techniques, classify and store them in the resource library and establish an index.

[0044] Virtual digital human debugging: Complete the construction of the virtual digital human model, input of the action library and integration of the knowledge graph, and test the naturalness of facial expressions, the smoothness of movements and the accuracy of question answering.

[0045] Virtual-Real Fusion Calibration: SLAM technology is used to complete the registration and calibration of virtual digital humans and real exhibition halls, and the lighting adaptation algorithm is adjusted to ensure natural fusion under different lighting conditions.

[0046] Interactive function testing: Test the response speed and recognition accuracy of three types of interaction methods: gesture, voice, and touch. Optimize priority scheduling rules to ensure that core interactions respond first.

[0047] System Deployment and Launch: Deploy the system to the exhibition hall terminals and online platform, open it to users, monitor the system's operating status in real time, and collect user feedback.

[0048] Iterative optimization: Based on user feedback and experience evaluation results, update resource library content, adjust system parameters, optimize interaction logic, and improve user experience.

[0049] III. Data Characterization for Effectiveness Verification Table 1: Performance Comparison of Suzhou Embroidery Technique AR Interactive Display System Before and After Implementation Evaluation indicators Before implementation (traditional demonstration) After implementation (of this system) Interactive engagement 35% 88% Knowledge of intangible cultural heritage 40% 82% User dwell time 5 minutes 25 minutes Content sharing rate 12% 65% User satisfaction 60% 91% Table 1 clearly demonstrates the application value of this system in showcasing Suzhou embroidery, an intangible cultural heritage. Interactive participation increased from 35% to 88%, primarily due to multimodal interaction and personalized storyline design. Users are no longer passive viewers but actively participate in the craft experience through gestures and voice, stimulating their enthusiasm. Knowledge acquisition of intangible cultural heritage increased from 40% to 82%. The virtual digital human accurately demonstrates needlework details and answers questions in real time, while knowledge points are embedded in the storyline, allowing users to quickly understand the core of Suzhou embroidery techniques. The dwell time increased from 5 minutes to 25 minutes, reflecting the richness of the system's content and the attractiveness of its interaction. Users can deeply experience different needlework techniques and browse historical and cultural resources. Content sharing rate increased from 12% to 65%, stemming from multi-person collaborative interaction and the experience video sharing function. Users are willing to share their interaction process on social media platforms, expanding the reach of Suzhou embroidery culture. User satisfaction reached 91%, reflecting the system's comprehensive performance in terms of natural integration of virtual and real elements, smooth interaction, and authoritative content, effectively addressing the pain points of traditional display formats being monotonous and lacking in user experience.

[0050] Example 2: Application of Online Dragon Boat Festival Intangible Cultural Heritage AR Interactive Display System This embodiment aims to promote the intangible cultural heritage of dragon boat racing during the Dragon Boat Festival. It provides online AR interactive experiences to users nationwide. Users can use mobile phones, tablets, and other terminals to summon virtual dragon boats and digital humans in real-world scenes such as riversides and parks near their homes. They can experience interactive content such as dragon boat making, racing, and folk rituals, thus solving the problem of geographical and time limitations in the dissemination of traditional dragon boat intangible cultural heritage.

[0051] I. System Deployment and Module Refinement The real-scene data acquisition module selects three typical dragon boat racing venues—rivers, lakes, and artificial lakes—for data collection. A 3D laser scanner scans the terrain, riverbank outlines, and water flow direction with a precision of 0.1 mm. A 4K high-definition camera captures scene images and lighting changes at different times of day—morning, noon, and evening—at a frame rate of 60 frames per second. A GPS positioning module calibrates the location of the venue. After gamma correction and noise reduction filtering, the collected data generates a 1:1 scale 3D point cloud model and high-definition texture map. This allows users to automatically match the terrain when using the data in similar real-world scenes, ensuring that the virtual dragon boat blends naturally with the real ground and water surface.

[0052] The Intangible Cultural Heritage Digital Resource Database for the Dragon Boat Festival is constructed, storing resources in four main categories: Dragon Boat Making, including animated breakdowns of traditional dragon boat construction steps such as wood selection, keel building, and painting decoration, 3D models of tools (planes, chisels, paintbrushes, etc.), and dimensional parameter documents; Racing Customs, including video materials and audio explanations of dragon boat racing team formation rules, paddling rhythms, and referee standards; Folk Rituals, including animated procedures for dragon boat eye-dotting, dragon worship, and dragon sending-off ceremonies, costume and prop models, and historical background text; and Interactive Storylines, including scripts for "Forming a Dragon Boat Team," "Preparing for the Race," and "Folk Performances," embedding dragon boat intangible cultural heritage knowledge points. All resources adopt standardized formats and are indexed using "region + category + keywords" to support rapid retrieval.

[0053] The virtual digital human construction module is modeled after dragon boat inheritors from southern China. It collects facial feature points, body posture data, and typical movements such as paddling and eye-dotting, using next-generation 3D modeling technology to construct a high-precision virtual digital human model. The face is bound to 52 muscle motion units, simulating natural expressions and demeanor such as exertion and cheers. Motion capture technology records paddling actions such as entering and exiting the water, eye-dotting, and dragon worship, down to the angle of arm force and the degree of body tilt, building a dedicated motion library containing over 1500 movements. Equipped with a natural language processing engine and a dragon boat intangible cultural heritage knowledge graph, it covers over 1200 knowledge points including dragon boat history, manufacturing techniques, and folk symbolism. It supports interaction in five dialects, including Mandarin, Cantonese, and Minnan, and can provide detailed explanations of the functions of each dragon boat component and racing techniques.

[0054] The AR fusion rendering module integrates SLAM technology to achieve accurate registration of the virtual dragon boat, digital human, and the user's surrounding real-world scene, with a registration error controlled within 2 pixels. It employs a lighting adaptation algorithm to dynamically adjust the lighting intensity, color temperature, and water reflection effects of the virtual digital human and dragon boat based on real-time ambient light data collected from the user's terminal, ensuring a natural blend of virtual and real elements. It supports adaptation to mobile phones and tablets, maintaining a rendering frame rate of 32 frames per second, supporting 4K output resolution, and allowing users to switch between "virtual-real overlay" and "pure virtual" display modes. The virtual digital human and scene adaptation optimization unit uses formulas... Calculate the virtual-real fusion adaptation degree, where The fit value is set to 0-100. The lighting adaptation weight is set to 0.3. This represents the illumination matching degree, with a value ranging from 0 to 100. The weight for proportional adaptation is set to 0.25, and S represents the proportional consistency degree, ranging from 0 to 100. The action is assigned a weight, with a value of 0.25. For motor coordination, the value ranges from 0 to 100. The environment adaptation weight is set to 0.2. Environmental fit, with a value ranging from 0 to 100. A user used it near a river at dusk, and it detected... Substituting into the formula, we get The system adjusts the reflectivity of the virtual digital human's clothing and the length of the dragon boat's shadow based on the calculation results, improving the fit to 93, which is highly consistent with the evening riverside scene.

[0055] The interactive module supports multimodal interaction and multi-person collaboration: the gesture recognition module captures the user's paddling movements with a recognition accuracy of 95%; when the user uses both hands to paddle, the virtual digital human exerts force synchronously, propelling the virtual dragon boat forward; the voice interaction module supports Mandarin and five dialects with a response time of 0.45 seconds; the system responds instantly to user commands such as "accelerate" and "turn left"; the touch interaction module supports clicking to switch folk rituals, swiping to adjust the dragon boat's direction, and zooming to view dragon boat details. The multi-person collaboration function supports 2-8 users entering the same AR scene via network connection. The system assigns independent virtual identities, supports real-time voice communication and collaborative paddling races, and generates interactive videos for sharing upon completion. The interactive response priority scheduling unit uses a formula... Determine the order of interaction request responses, where This sets the priority for interactive responses, with a value ranging from 1 to 10. As a weight for timeliness, a value of 0.3 is assigned. The urgency level of the request is indicated by a value from 1 to 10. This is the content importance weight, with a value of 0.25. The value of the interactive content intangible cultural heritage is rated from 1 to 10. The weight for user demand is set to 0.25. User activity level, with a value from 1 to 10. This represents the system load weight, with a value of 0.2. This represents the current system resource utilization rate, ranging from 1 to 10. A user initiates a "dotting the eyes" ceremony request during a multi-player dragon boat race. =9、 =10、 =8、 =6, substituting into the formula, we get =0.3×9+0.25×10+0.25×8+0.2×6=2.7+2.5+2+1.2=8.4. This priority is higher than the ordinary paddling command, and the system will trigger the eye-dotting ceremony animation first to ensure that the core folk interaction is not interrupted.

[0056] The scene adaptation optimization and data storage management modules monitor the ambient light, terminal performance, and network bandwidth in real time: In strong outdoor light environments, the brightness of virtual objects is automatically reduced to avoid overexposure; when terminal performance is insufficient, the polygon count of the dragon boat model is reduced from 1.2 million to 900,000; when network bandwidth is below 10Mbps, incremental transmission is used, loading the dragon boat body and core actions first, then adding details such as painted textures. For complex outdoor environments, anti-interference algorithms are added to ensure SLAM registration stability and avoid fusion offset caused by scene occlusion. The data storage management module adopts a blockchain + distributed database architecture. The blockchain stores key data such as user interaction records, racing results, and scene creation logs; the distributed database stores scene acquisition data, virtual models, and resource library data, supporting 1300 data read / write operations per second. AES-256 encryption technology ensures data security, and three-level access permissions distinguish between administrators, ordinary users, and visitors. The data backup mechanism includes daily incremental backups and weekly full backups, and also supports users saving interaction records and experience videos locally.

[0057] Other functional modules' user experience evaluation unit uses formulas. Quantify user experience effects, among which This is a comprehensive score for user experience, ranging from 0 to 100. To incorporate the smoothness weight, a value of 0.3 is used. The smoothness of virtual-real integration is scored, with a value of 0-100. The weight for interactive immersion is set to 0.25. Rate the level of immersion in the interactive experience, on a scale of 0-100. The content satisfaction weight is set to 0.25. Rate the satisfaction with intangible cultural heritage content, with a value of 0-100. As a weight for learning effectiveness, a value of 0.2 is assigned. The learning effectiveness of intangible cultural heritage knowledge is scored, ranging from 0 to 100. Statistics are compiled after a particular operation. =89、 =90、 =91、 =86, substituting into the formula, we get... =0.3×89+0.25×90+0.25×91+0.2×86=26.7+22.5+22.75+17.2=89.15. Based on the results, the system optimizes the multi-user collaborative rendering algorithm and voice interaction recognition model to improve the user experience. The resource update unit collaborates with dragon boat intangible cultural heritage protection institutions in various regions to obtain characteristic data on dragon boats from different regions, such as Hunan and Guangdong, and updates it to the resource library after expert review. The scene expansion module's editor allows users to import custom site photos, add dragon boat decorations and folk elements, set racing rules, and invite friends to participate after publishing.

[0058] II. System Implementation Process Scene acquisition and resource integration: Complete the scanning and data processing of typical dragon boat racing venues, collect and organize digital resources of dragon boat intangible cultural heritage from different regions, build a resource library and establish an index.

[0059] Virtual Digital Human and Dragon Boat Debugging: Complete the construction of virtual digital human models, motion libraries and knowledge graphs, build 3D models of different types of dragon boats, and test the smoothness of movements and the effect of virtual-real integration.

[0060] Interactive feature development: Develop gesture, voice, and touch interaction functions and multi-person collaboration modules, debug priority scheduling rules, and ensure that core interactions respond first.

[0061] Adaptability testing: Test the system's operating status under different lighting, terminal, and network conditions, optimize the scene adaptation algorithm, and ensure stable operation.

[0062] Launch, promotion, and operation: Launch the system on the application platform, carry out Dragon Boat Festival-themed promotional activities, collect user feedback, and monitor system operation in real time.

[0063] Iterative Updates: Based on user feedback and experience evaluation results, the resource library is updated, system performance is optimized, and regionally distinctive dragon boats and interactive scenes are added.

[0064] III. Data Characterization for Effectiveness Verification Table 2: Performance Comparison of the Dragon Boat Festival Intangible Cultural Heritage AR Interactive Display System Before and After Implementation Evaluation indicators Before implementation (traditional communication) After implementation (of this system) Geographical coverage Local areas Nationwide Number of participants More than a thousand people participated in a single event. Hundreds of thousands of people in a single event Depth of understanding of folk customs superficial understanding Deep cognition Event duration Dragon Boat Festival Available year-round User sharing intentions 20% 70% Table 2 data highlights the breakthroughs this system has made in the dissemination of dragon boat intangible cultural heritage. Geographical coverage has expanded from localized areas to the entire country, breaking the traditional limitations of venue and location in dragon boat racing. Users across the country can experience the sport in their immediate surroundings without needing to travel to specific locations, significantly expanding the audience. The number of participants has increased from a few thousand per event to hundreds of thousands, thanks to the convenience of online dissemination and the appeal of interactive experiences. The multi-person collaborative racing function further enhances user participation. The depth of folk custom understanding has shifted from superficial to in-depth. The system uses virtual digital humans to demonstrate details such as dragon boat making and the eye-dotting ceremony, combined with real-time explanations using knowledge graphs, allowing users to fully grasp the cultural connotations and craftsmanship of dragon boat intangible cultural heritage. The duration of the activity has been extended from the Dragon Boat Festival to year-round, changing the traditional situation of "high time sensitivity and short dissemination cycle" in intangible cultural heritage dissemination. Users can experience the sport anytime, deepening their memory. The user willingness to share increased from 20% to 70%, thanks to the interactive video sharing function and the social attributes of multi-person collaborative experience. Users are happy to share their experience, forming secondary dissemination, further expanding the cultural influence of dragon boat intangible cultural heritage, and achieving the dual goals of living inheritance and wide dissemination.

[0065] Reference Figure 2This diagram visually demonstrates the advantages of this system in fusion of virtual and real elements across multiple scenarios. Traditional AR systems lack optimization for lighting, scale, motion, and environment, only achieving simple overlay, with adaptation scores generally below 55%, resulting in a strong disconnect between virtual objects and real-world scenes. This system, through precise SLAM registration, lighting adaptation algorithms, and a fusion adaptation calculation formula, dynamically adjusts the parameters of the virtual digital human, achieving adaptation scores exceeding 85%, and reaching 93 points in the riverside scene. This stems from the system's accurate capture and real-time adaptation of terrain, lighting, and environmental features in different scenes, ensuring that the virtual digital human can naturally blend into various real-world environments.

[0066] Reference Figure 3 This diagram comprehensively reflects the system's overall advantages in user experience. Traditional display methods mainly rely on static displays and text explanations, with a smoothness and interactive immersion score of less than 45 points, making it difficult for users to gain a deep understanding of intangible cultural heritage content. This system scores over 85 points in all dimensions, with a smoothness score of 90 points due to the natural and harmonious integration of virtual and real elements; an interactive immersion score of 88 points, stemming from multimodal interaction and personalized storylines; a content satisfaction score of 91 points, thanks to the precise skill demonstrations and authoritative knowledge explanations provided by virtual digital humans; a learning effectiveness score of 85 points, due to the embedding of knowledge points and real-time Q&A within the storylines; and an ease of operation score of 89 points, adaptable to multiple terminals and with simple interactive logic.

[0067] Reference Figure 4 This diagram highlights the system's high efficiency in disseminating intangible cultural heritage. Traditional methods are limited by geography and time, relying on offline activities and word-of-mouth, reaching only 50,000 people in 30 days, resulting in a narrow reach. This system, deployed across multiple online terminals, supports nationwide users anytime, anywhere. Combined with multi-user collaborative interaction and video sharing features, dissemination has grown exponentially, reaching 850,000 people in 30 days. This is because the system breaks down geographical barriers to intangible cultural heritage dissemination, allowing users to experience it in their surrounding real-world settings, and the sharing function facilitates secondary dissemination. Simultaneously, the accurate recreation of intangible cultural heritage skills by virtual digital humans allows more people to easily access and understand intangible cultural heritage, effectively expanding the audience and contributing to the widespread dissemination and living transmission of intangible cultural heritage.

[0068] The above are merely preferred embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A non-material cultural heritage AR interactive display system that integrates virtual digital humans with real-world scenes, characterized in that, Includes the following modules: Real-world scene acquisition module: Equipped with a 3D laser scanner, camera and GPS positioning module, it acquires the 3D coordinates, topography, exhibit parameters, scene images and dynamic images of the intangible cultural heritage display scene; it performs gamma correction, noise reduction filtering and image stitching processing on the acquired data to generate a 1:1 scale 3D point cloud model and texture map; The Intangible Cultural Heritage Digital Resource Database adopts a distributed storage architecture to classify and store digital resources of intangible cultural heritage in the categories of traditional skills, performances, and folk festivals; the resources adopt a standardized format and are indexed by keywords, intangible cultural heritage categories, and regional attributes; Virtual Digital Human Construction Module: Collects facial features, body posture and motion data, uses next-generation 3D modeling technology to construct a digital human model, binds 52 muscle units to the face, records skill demonstrations and interactive postures through motion capture technology, builds a dedicated motion library, and is equipped with a natural language processing engine and intangible cultural heritage knowledge graph; AR fusion rendering module: integrates SLAM technology to realize digital human registration in virtual and real scenes, and coordinates virtual and real lighting and shadows through lighting adaptation algorithms; Interactive module: Supports gesture, voice, and touch interaction; voice recognition covers Mandarin and 20+ dialects; Touchscreen compatible with mobile devices, supporting click and swipe operations; Scene adaptation and optimization module: Real-time monitoring of ambient light, device performance and network bandwidth; night mode is enabled when light is ≤500 lux; model accuracy is reduced when device performance is insufficient; incremental transmission is used when bandwidth is ≤10Mbps. Data storage management module: Blockchain stores key data; distributed database stores raw data of scenario collection, models and resources; AES-256 encryption is used to set access permissions and establish a backup mechanism.

2. The AR interactive display system for intangible cultural heritage that integrates virtual digital humans and real scenes according to claim 1, characterized in that, It also includes a virtual digital human and scene adaptation optimization unit, through Calculate the virtual-real fusion adaptation degree, where F is the fusion adaptation degree value. To adapt weights to lighting conditions, For lighting matching degree, To adapt the weights proportionally, For the sake of proportional harmony, To adapt weights to actions, For motor coordination, For environment adaptation weights, For environmental compatibility.

3. The AR interactive display system for intangible cultural heritage that integrates virtual digital humans and real scenes according to claim 1, characterized in that, It also includes an interactive story generation module: an interactive story framework is built based on the core content of intangible cultural heritage, including the main story and side story. The system dynamically generates customized story based on user interaction behavior, selection preferences and dwell time. Interactive links are set at story nodes, and the results of the interaction affect the direction of the story. Explanations of intangible cultural heritage knowledge points are embedded in the story.

4. The AR interactive display system for intangible cultural heritage that integrates virtual digital humans and real scenes according to claim 1, characterized in that, It also includes an interactive response priority scheduling unit, through Determine the order of interaction request responses, where Prioritize interactive responses. In response to timeliness weighting, To request the urgency level, As a weight for content importance, To enhance the intangible cultural heritage value of interactive content, Weighting based on user demand. To increase user activity As the system load weight, This represents the current system resource utilization rate; the higher the priority value, the faster the response.

5. The AR interactive display system for intangible cultural heritage that integrates virtual digital humans and real scenes according to claim 1, characterized in that, The virtual digital human construction module also includes an intelligent learning unit: by collecting user interaction feedback data, the latest research results on intangible cultural heritage, and updated content on the skills of inheritors, deep learning algorithms are used to optimize the language expression, action demonstration, and knowledge reserves of the virtual digital human. The virtual digital human action library and knowledge graph are updated every quarter, and the model accuracy is iterated and upgraded once a year.

6. The AR interactive display system for intangible cultural heritage that integrates virtual digital humans and real scenes according to claim 1, characterized in that, It also includes a user experience evaluation unit, through Quantify user experience effects, among which The overall score for user experience To incorporate smoothness weights, The smoothness of the virtual-real integration is scored. Weighting for interactive immersion, Rate the level of interactive immersion. As a weight for content satisfaction, Rate the satisfaction with intangible cultural heritage content. Weighting of learning outcomes The system scores the learning outcomes of intangible cultural heritage knowledge and dynamically adjusts system parameters based on the evaluation results.

7. The AR interactive display system for intangible cultural heritage that integrates virtual digital humans and real scenes according to claim 1, characterized in that, The AR fusion rendering module also integrates a virtual-real collision detection unit: it uses the AABB bounding box algorithm to construct collision objects between virtual digital humans, virtual props and real scenes. The collision detection frequency is consistent with the rendering frame rate. When the distance between the virtual object and the real scene physical exhibit is less than 5 centimeters, the collision response mechanism is automatically triggered to adjust the position or transparency of the virtual object.

8. The AR interactive display system for intangible cultural heritage that integrates virtual digital humans and real scenes according to claim 1, characterized in that, The Intangible Cultural Heritage Digital Resource Database also includes a resource update unit: establishing a dynamic update mechanism for intangible cultural heritage resources, setting up data connection channels with intangible cultural heritage protection centers, inheritors and academic research institutions, obtaining the latest intangible cultural heritage skill videos, documents and inheritor stories in real time, reviewing and confirming resources before updates, and notifying users through the system push function after updates, guiding users to experience the new content, retaining historical version resources, and supporting users to view and compare.

9. The AR interactive display system for intangible cultural heritage that integrates virtual digital humans and real scenes according to claim 1, characterized in that, The interactive module also supports multi-user collaborative interaction: multiple users enter the same AR interactive scene through network connection, the system assigns independent virtual identity identifiers, and distributed rendering technology is used during multi-user interaction to record the multi-user interaction process and generate interactive videos, which can be shared and saved by users to expand the scope of intangible cultural heritage dissemination.

10. The AR interactive display system for intangible cultural heritage that integrates virtual digital humans and real scenes according to claim 1, characterized in that, It also includes a scene expansion module: it supports users to customize intangible cultural heritage AR display scenes, provides a scene editor tool, allows users to import custom scene models, add virtual props and set interaction rules, the scene editor has a built-in intangible cultural heritage element material library, supports drag-and-drop operation and parameter adjustment, and after the custom scene is generated, it is published after being reviewed by the system for other users to experience.

Images