Digital exhibition hall system and visitor guiding method thereof

Abstract

The invention relates to a digital exhibition hall system and a visitor guiding method thereof.The method includes the steps that a 2D digital person is generated through a 2D digital person front-end assembly to communicate with a visitor in a dialogue mode, a visitor dialogue text stream is output to an intention recognition model, and the intention recognition model continuously monitors the dialogue text stream and outputs an intention label; the trigger strategy engine judges whether 3D guidance is triggered or not according to the intention label, and when the 3D guidance is triggered, the 3D digital person guiding system generates a personalized route and a parastyle script, and guides a visitor through a 3D digital person; visitors are received through the 2D digital person, and skipping to the 3D digital person is carried out according to the intention of the visitors, so that 2D chat is accurately switched to the 3D exhibition hall, and meanwhile, the situation that the visitors are limited to a plane chat interface, so that immersion is lacked, the staying time of the visitors is prolonged, low-value traffic still stays in a 2D link, and 3D rendering resources are only open to high potential customers is avoided. The computing power of the server is saved, and the cost is saved.

Description

Technical Field

[0001] This application relates to the field of digital exhibition hall technology, specifically to a digital exhibition hall system that integrates 2D digital humans and 3D digital humans, and a visitor guidance method thereof. Background Technology

[0002] Pure 2D digital human online reception systems typically consist of 2D virtual customer service embedded in web pages or apps, using TTS+NLP technology to answer visitor questions. Common products include domestic and international intelligent customer service systems (Xiaoice, Udesk digital humans, etc.). 3D metaverse exhibition halls combined with NPC digital guides are often collaborations between large enterprises and cloud exhibition platforms to build highly immersive 3D virtual exhibition halls. Visitors navigate using a mouse / VR, with 3D virtual humans providing narration. However, initial lead generation mainly relies on QR codes or external links. Conversational marketing funnel SaaS, on the other hand, uses pop-up windows on e-commerce / official website homepages to recommend products and collect leads, primarily focusing on form conversion and lacking a virtual space experience.

[0003] The aforementioned reception system has the following problems:

[0004] 2D digital human conversion rate is low: flat chat interface lacks immersion, visitors stay for a short time and the depth of interaction is limited; human intervention is still required after multiple rounds of dialogue.

[0005] 3D digital exhibition halls are costly and have scattered entrances: they need to be built and maintained separately, and only high-intent customers can see them; many potential customers are stopped at 2D pages and cannot enjoy the advantages of 3D scenes.

[0006] Cross-scene connection gap: 2D chat to 3D digital exhibition hall usually relies on passive links or manual invitations, resulting in a high bounce rate.

[0007] Lack of intelligent intent recognition: Existing systems cannot accurately determine whether visitors are worth guiding into the 3D space in the early stages of a conversation, resulting in either interrupting the experience by "force-ing" them in or missing business opportunities. Summary of the Invention

[0008] In view of the above problems, this application provides a digital exhibition hall system that integrates 2D digital humans and 3D digital humans and its visitor guidance method, which solves the problems of low conversion rate of existing 2D digital humans, high cost of 3D digital exhibition halls, scattered entrances and disconnection between cross-scenes.

[0009] To achieve the above objectives, the inventors provide a digital exhibition hall system that integrates 2D and 3D digital humans, comprising:

[0010] The 2D digital human front-end component is used to communicate with visitors through a 2D digital human and output the visitor dialogue text stream. When it receives the packaged information pushed back from the 3D exhibition hall access gateway, it jumps to the 3D digital human tour guide system.

[0011] The intent recognition module is used to perform intent recognition on the dialogue text stream output by the 2D digital human front-end component and output intent tags.

[0012] The trigger strategy engine is used to determine whether to trigger 3D guidance based on the intent label output by the intent recognition module. When it is determined that 3D guidance is triggered, the visitor is guided into the 3D digital exhibition hall.

[0013] The 3D exhibition hall access gateway is used to package the visitor's identity, contextual intent, and product preferences and push them to the 2D digital human front-end component after determining that a user has entered the 3D digital exhibition hall.

[0014] A 3D digital human tour guide system is used to generate personalized routes and narration scripts based on the visitor's identity, contextual intent, and product preferences. The system uses 3D digital humans to receive visitors and renders 3D digital exhibition halls in real time.

[0015] In some embodiments, the triggering strategy engine is specifically used to determine whether the intent tag is greater than a set threshold, the dwell time is greater than a preset time, and the keyword weight is greater than a preset weight. If so, it determines to trigger 3D guidance.

[0016] In some embodiments, the triggering strategy engine is specifically used to guide visitors into the 3D digital exhibition hall by popping up a 3D digital exhibition hall guide card or button.

[0017] In some embodiments, the 3D showroom access gateway is used to package visitor identity, contextual intent, and product preferences into a URL / QR-Code format and push it to the 2D digital human front-end component by generating a one-time token and deep chain parameters.

[0018] In some embodiments, it also includes:

[0019] The data attribution and feedback module records the switching rate from 2D digital human to 3D digital human, visitor dwell time, and completed orders, and feeds this information back to the intent recognition model and trigger strategy engine.

[0020] Another technical solution is also provided: a visitor guidance method for a digital exhibition hall system that integrates 2D and 3D digital humans, including the following steps:

[0021] The 2D digital human front-end component generates a 2D digital human to communicate with visitors and outputs the visitor's dialogue text stream.

[0022] The intent recognition module performs intent recognition on the dialogue text stream output by the 2D digital human front-end component and outputs intent labels;

[0023] The trigger strategy engine determines whether to trigger 3D guidance based on the intent tags output by the intent recognition module. If it determines that 3D guidance is triggered, it guides visitors into the 3D digital exhibition hall.

[0024] Once the 3D exhibition hall access gateway confirms that a user has entered the 3D digital exhibition hall, it pushes the visitor's identity, contextual intent, and product preferences to the 2D digital human front-end component.

[0025] The 3D digital human guide system generates personalized routes and narration scripts based on the user's visitor identity, contextual intent, and product preferences. It uses 3D digital humans to greet visitors and renders 3D digital exhibition halls in real time.

[0026] In some embodiments, the triggering strategy engine determines whether to trigger 3D guidance based on the intent label output by the intent recognition module, specifically including the following steps:

[0027] The trigger strategy engine determines whether to trigger 3D guidance by judging whether the intent tag is greater than the set threshold, the dwell time is greater than the preset time, and the keyword weight is greater than the preset weight.

[0028] In some embodiments, guiding visitors into the 3D digital exhibition hall specifically includes the following steps:

[0029] The trigger strategy engine guides visitors into the 3D digital exhibition hall by popping up a guide card or button.

[0030] In some embodiments, the step of packaging visitor identity, contextual intent, and product preferences and pushing them to the 2D digital human front-end component specifically includes the following steps:

[0031] The 3D showroom access gateway generates a one-time token and deep chain parameters, packages the visitor's identity, contextual intent, and product preferences, and pushes them to the 2D digital human front-end component in the form of a URL / QR-Code.

[0032] In some embodiments, the following steps are also included:

[0033] The data attribution and feedback module records the switching rate from 2D digital human to 3D digital human, visitor dwell time, and completed orders, and feeds this information back to the intent recognition model and trigger strategy engine.

[0034] Unlike existing technologies, the above technical solution, when receiving visitors, first generates a 2D digital human through the 2D digital human front-end component to communicate with the visitor, handling the initial dialogue, FAQs (Frequently Asked Questions), and information push. It then outputs the visitor's dialogue text stream to the intent recognition model. This intent recognition model, based on a multi-task converter trained on a large-scale labeled dialogue corpus, continuously monitors the dialogue text stream, outputs intent tags, and triggers a strategy engine. Based on the intent tags output by the intent recognition module, it determines whether to trigger 3D guidance. When 3D guidance is triggered, the 3D showroom access gateway pushes the visitor's identity, contextual intent, and product preferences to the 2D digital human front-end component. Upon receiving the packaged information pushed back from the 3D showroom access gateway, the 2D digital human front-end component redirects to the 3D showroom. The 3D digital human guide system generates personalized routes and narration scripts based on visitor identity, context, clothing, and product preferences. It guides visitors through a 3D digital human and renders a 3D digital exhibition hall in real time. Visitors are greeted by a 2D digital human, and the system automatically transitions them to the 3D digital human based on their intentions. This allows for precise switching from 2D chat to the 3D exhibition hall, avoiding a lack of immersion due to limited interaction with a flat chat interface, extending visitor dwell time, and keeping low-value traffic in the 2D environment. 3D rendering resources are only available to high-potential visitors, saving server computing power and thus reducing costs.

[0035] The above description of the invention is merely an overview of the technical solution of this application. In order to enable those skilled in the art to better understand the technical solution of this application and to implement it based on the description and drawings, and to make the above-mentioned objectives and other objectives, features and advantages of this application easier to understand, the following description is provided in conjunction with the specific embodiments and drawings of this application. Attached Figure Description

[0036] The accompanying drawings are only used to illustrate the principles, implementation methods, applications, features, and effects of specific embodiments of this application and other related content, and should not be considered as limitations on this application.

[0037] In the accompanying drawings of the instruction manual:

[0038] Figure 1 This is a schematic diagram of a digital exhibition hall system that integrates 2D and 3D digital humans, as described in a specific implementation.

[0039] Figure 2 This is a schematic diagram of another structure of the digital exhibition hall system integrating 2D and 3D digital humans described in a specific implementation;

[0040] Figure 3This is a schematic diagram of another structure of the digital exhibition hall system integrating 2D and 3D digital humans described in a specific implementation;

[0041] Figure 4 This is a schematic diagram illustrating the implementation process of the digital exhibition hall system that integrates 2D and 3D digital humans, as described in a specific embodiment.

[0042] Figure 5 This is a flowchart illustrating a visitor guidance method for a digital exhibition hall system that integrates 2D and 3D digital humans, as described in a specific implementation.

[0043] The reference numerals used in the above figures are explained as follows:

[0044] 100. 2D digital human front-end components.

[0045] 200. Intent recognition module.

[0046] 300. Trigger the strategy engine.

[0047] 400, 3D exhibition hall access gateway,

[0048] 500, 3D digital human guide system,

[0049] 600. Data Attribution and Feedback Module. Detailed Implementation

[0050] To illustrate the possible application scenarios, technical principles, implementable specific solutions, and achievable objectives and effects of this application in detail, the following description, in conjunction with the listed specific embodiments and accompanying drawings, provides a detailed explanation. The embodiments described herein are merely illustrative of the technical solutions of this application and are therefore intended to limit the scope of protection of this application.

[0051] In this document, the term "embodiment" means that a specific feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The term "embodiment" appearing in various places throughout the specification does not necessarily refer to the same embodiment, nor does it specifically limit its independence or connection with other embodiments. In principle, in this application, as long as there are no technical contradictions or conflicts, the technical features mentioned in each embodiment can be combined in any way to form corresponding implementable technical solutions.

[0052] Unless otherwise defined, the technical terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains; the use of related terms herein is merely for the purpose of describing particular embodiments and is not intended to limit this application.

[0053] In the description of this application, the term "and / or" is used to describe the logical relationship between objects, indicating that three relationships can exist. For example, A and / or B means: A exists, B exists, and A and B exist simultaneously. Additionally, the character " / " in this document generally indicates that the preceding and following objects have an "or" logical relationship.

[0054] In this application, terms such as “first” and “second” are used only to distinguish one entity or operation from another, and do not necessarily require or imply any actual quantity, hierarchy or order relationship between these entities or operations.

[0055] Unless otherwise specified, the use of terms such as “comprising,” “including,” “having,” or other similar expressions in this application is intended to cover non-exclusive inclusion, which does not exclude the presence of additional elements in a process, method, or product that includes the stated elements, such that a process, method, or product that includes a list of elements may include not only those defined elements but also other elements not expressly listed, or elements inherent to such a process, method, or product.

[0056] Similar to the understanding in the Examination Guidelines, in this application, expressions such as "greater than," "less than," and "exceeding" are understood to exclude the stated number; expressions such as "above," "below," and "within" are understood to include the stated number. Furthermore, in the description of the embodiments in this application, "multiple" means two or more (including two), and similar expressions related to "multiple" are also understood in this way, such as "multiple groups" and "multiple times," unless otherwise explicitly specified.

[0057] In the description of the embodiments of this application, the space-related expressions used, such as "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "vertical," "top," "bottom," "inner," "outer," "clockwise," "counterclockwise," "axial," "radial," and "circumferential," indicate the orientation or positional relationship based on the orientation or positional relationship shown in the specific embodiments or drawings. They are only for the purpose of describing the specific embodiments of this application or for the reader's understanding, and do not indicate or imply that the device or component referred to must have a specific position, a specific orientation, or be constructed or operated in a specific orientation. Therefore, they should not be construed as limitations on the embodiments of this application.

[0058] Unless otherwise expressly specified or limited, the terms "installation," "connection," "linking," "fixing," and "setting," as used in the description of the embodiments of this application, should be interpreted broadly. For example, "connection" can be a fixed connection, a detachable connection, or an integral setting; it can be a mechanical connection, an electrical connection, or a communication connection; it can be a direct connection or an indirect connection through an intermediate medium; it can be the internal connection of two components or the interaction between two components. For those skilled in the art to which this application pertains, the specific meaning of the above terms in the embodiments of this application can be understood according to the specific circumstances.

[0059] Please see Figure 1 This embodiment provides a digital exhibition hall system that integrates 2D and 3D digital humans, including:

[0060] The 2D digital human front-end component 100 is used to communicate with visitors through a 2D digital human and output the visitor dialogue text stream. When it receives the packaged information pushed back by the 3D exhibition hall access gateway, it jumps to the 3D digital human tour guide system.

[0061] The intent recognition module 200 is used to perform intent recognition on the dialogue text stream output by the 2D digital human front-end component 100 and output intent tags.

[0062] Triggering strategy engine 300 is used to determine whether to trigger 3D guidance based on the intent label output by intent recognition module 200. When it is determined that 3D guidance is triggered, the visitor is guided into the 3D digital exhibition hall.

[0063] The 3D exhibition hall access gateway 400 is used to push visitor identity, contextual intent and product preferences to the 2D digital human front-end component 100 after determining that a user has entered the 3D digital exhibition hall.

[0064] A 3D digital human tour guide system 500 is used to generate personalized routes and narration scripts based on the visitor's identity, contextual intent, and product preferences, and to receive visitors through 3D digital humans and render 3D digital exhibition halls in real time.

[0065] When receiving visitors, the 2D digital human front-end component 100 first generates a 2D digital human to communicate with the visitor, handling the initial dialogue, FAQs (Frequently Asked Questions), and information push. It also outputs the visitor's dialogue text stream to the intent recognition model. This intent recognition model, trained on a large-scale labeled dialogue corpus, continuously monitors the dialogue text stream, outputs intent tags, and triggers the strategy engine 300. Based on the intent tags output by the intent recognition module 200, it determines whether to trigger 3D guidance. When 3D guidance is triggered, the 3D showroom access gateway 400 pushes the visitor's identity, contextual intent, and product preferences to the 2D digital human front-end component 100. When the 2D digital human front-end component 100 receives the packaged information pushed back from the 3D showroom access gateway 400... Then, visitors are redirected to the 3D digital human guide system 500. Based on the visitor's identity, context, clothing, and product preferences, the 3D digital human guide system 500 generates personalized routes and narration scripts, guiding visitors through the 3D digital human and rendering the 3D digital exhibition hall in real time. Visitors are greeted by a 2D digital human and are redirected to the 3D digital human according to their intentions. This achieves precise switching from 2D chat to the 3D exhibition hall, while avoiding a lack of immersion caused by visitors being limited to a flat chat interface, extending visitor dwell time, and keeping low-value traffic in the 2D stage. 3D rendering resources are only available to high-potential customers, saving server computing power and thus saving costs.

[0066] In some embodiments, the triggering strategy engine 300 is specifically used to determine whether the intent tag is greater than a set threshold, the dwell time is greater than a preset time, and the keyword weight is greater than a preset weight. If so, it determines to trigger 3D guidance.

[0067] The system determines whether 3D guidance needs to be triggered by judging whether the intent tag is greater than the set threshold, the dwell time is greater than the preset time, and the keyword weight. If the intent tag is greater than the set threshold, the dwell time is greater than the preset market, and the keyword weight is greater than the preset weight, it is determined that the visitor needs in-depth interaction, and 3D guidance is triggered to guide the visitor to the 3D exhibition hall.

[0068] In other embodiments, the determination of whether to trigger 3D guidance can also be made by judging whether the intent tag is greater than a set threshold, the dwell time is greater than a preset time, and the keyword weight, among one or more of these conditions.

[0069] In some embodiments, the triggering strategy engine 300 is specifically used to guide visitors into the 3D digital exhibition hall by popping up a 3D digital exhibition hall guide card or button.

[0070] Visitors are guided into the 3D digital exhibition hall by pop-up 3D digital exhibition hall guide cards or buttons. Customers can click on the 3D digital exhibition hall guide cards or buttons to confirm whether they want to enter the 3D digital exhibition hall.

[0071] In other embodiments, visitors can be asked by voice whether they need to enter the 3D digital exhibition hall.

[0072] In some embodiments, the 3D showroom access gateway 400 is used to package visitor identity, contextual intent, and product preferences by generating a one-time token and deep chain parameters, and push them to the 2D digital human front-end component 100 in the form of a URL / QR-Code.

[0073] A One-Time Token (OTT) is a one-time password used for identity verification or authorization. It expires immediately after use and cannot be reused. Its extremely short validity period and automatic expiration upon expiration, coupled with an algorithm-generated unpredictable random value, prevent guessing or replay attacks and ensure visitor information is not leaked. Deep Link Parameters are key-value pairs embedded in deep links, used to pass specific data to the target application when a user clicks the link, enabling precise page redirection or personalized operations. A URL (Uniform Resource Locator) is an address standard used on the internet for locating and accessing resources; it is a common form of URI (Uniform Resource Identifier). A QR Code (Quick Response Code) is a matrix-style two-dimensional barcode capable of storing large amounts of data and being read quickly.

[0074] Please see Figure 2 In some embodiments, it also includes:

[0075] The data attribution and feedback module 600 is used to record the switching rate from 2D digital human to 3D digital human, visitor dwell time and order completion, and feeds it back to the intent recognition model and trigger strategy engine 300.

[0076] By recording metrics such as the switching rate from 2D digital human to 3D digital human, visitor dwell time, and number of orders, the data is fed back to the intent recognition model and trigger strategy engine 300 for reverse updates, making intent recognition more accurate and determining whether to provide 3D guidance more precise.

[0077] In some embodiments, the function of the 3D exhibition hall access gateway 400 pushing the packaged message back to the 2D digital human front-end component 100 is as follows:

[0078] 1. State synchronization and session continuity;

[0079] The pushed-back data packet contains visitor_id, session context summary, exhibition hall deep link URL / QR-Code, one-time token, and timeout information.

[0080] Once the 2D digital human obtains these fields in real time, it can generate a "Continue Tour" card or rich media button in the original dialog window, and display a real-time progress bar and exit / return button on the UI to visitors after entering the 3D space, thus achieving uninterrupted cross-dimensional conversations.

[0081] 2. Security verification and access control;

[0082] The 2D client uses the token to verify the visitor's identity again; when the visitor scans the code to enter the 3D scene on another terminal (such as a mobile phone), the 2D client automatically switches to "read-only spectator" mode to prevent the same token from being reused or intercepted.

[0083] 3. Failure rollback and fallback experience;

[0084] If 3D rendering fails (due to low-end devices / weak network) or the visitor is interrupted during the loading phase, the 2D client can display a 2D text / image / video version of the same content based on the fallback_plan field, ensuring complete information delivery.

[0085] 4. Data attribution closed loop;

[0086] Visitor behavior events in the 3D space (hotspot clicks, dwell time, favorites list, etc.) are incrementally written back to the hidden channel on the 2D end via the gateway WebSocket in real time, for use in immediate customer service intervention or CRM lead scoring.

[0087] In some embodiments, the 3D digital human tour guide system 500 loads visitor context, generates personalized routes and narration scripts, and renders them in real time using WebGL / WebXR as follows:

[0088] 1. Based on Context Loader, obtain visitor_context (FAQ records, click logs, and interest vectors) through the access gateway REST API. Write it to the Session Graph (Redis).

[0089] 2. Based on Interest Profiler, run an interest extraction model based on Seq2Seq+Attention, score the context, and output a Top-K list of {product line: interest level}.

[0090] 3. Based on Path Planner, perform multi-objective A* search on the exhibition hall Scene Graph (directed weighted graph):

[0091] • Heuristic functions combine interest level, item distance, and congestion level;

[0092] • Constraints: Total duration ≤ 5 minutes; route can be rewound;

[0093] Output node sequence, e.g., ['Entry', 'Product A Area', 'Solution B Platform', 'Demo Area', 'Feedback Terminal'].

[0094] 4. Based on Narration Generator, call GPT-x to fine-tune the model for each node: prompt = industry + visitor intent + exhibit metadata, generate Chinese narration script, and then pass it through the text review rule engine (banned words & length limits).

[0095] 5. Based on Asset Preloader, batch fetch 3D Mesh, textures, LOD resources, and audio according to node sequence; utilizing... <code> requestIdleCallback< / code> &WebWorker performs multi-threaded decompression to prevent the main thread from being blocked.

[0096] 6. Based on WebGL Renderer, initialize the Three.js scene, PBR materials, and HDR environment maps; enable GPU Instancing and Frustum Culling; the digital human skeletal animation uses the GPU Skinning pipeline.

[0097] 7. Based on WebXR Adaptor (optional): If the browser supports WebXR and the device is a VR / AR headset, then:

[0098] • Switch to XR session and update the camera stereo matrix;

[0099] • Enable foveated rendering.

[0100] 8. Based on Realtime Sync & UI, digital human lip-sync: using Visem-Driven BlendShape, audio frame → visem → Morph Target;

[0101] —The navigation UI (route points, hotspot highlights) uses a dual-layer rendering of HTML Overlay and 3D tags.

[0102] 9. Based on the Event Logger, events such as node_enter, node_exit, and cta_click are updated in real time via the WebSocket → Data Attribution module to determine the weights of the intent model.

[0103] Please see Figure 3 In some embodiments, a digital exhibition hall system integrating 2D and 3D digital humans is provided, including:

[0104] 2D Digital Human Front-End Component 100: Handles initial dialogue, FAQ, and data push; outputs visitor dialogue text stream.

[0105] Intent recognition model: a multi-task Transformer trained on a large-scale labeled dialogue corpus (industry knowledge, historical conversion samples); continuously listens to the dialogue flow and outputs intent labels P (intent|context).

[0106] Trigger Strategy Engine 300: Set thresholds and contextual conditions (such as visitor browsing time, keyword weight) to determine whether to trigger 3D guidance.

[0107] 3D showroom access gateway 400: Generate a one-time token and deep chain parameters, package visitor identity, contextual intent, and product preferences, and push them in the form of URL / QR-Code.

[0108] 3D Digital Human Navigation System 500: Loads visitor context, personalized routes and narration scripts; renders in real time via WebGL / WebXR.

[0109] Data Attribution and Feedback Module 600: Records metrics such as 2D→3D switching rate, dwell time, and number of orders; updates intent models and trigger strategies in reverse.

[0110] Please see Figure 4 The implementation process is as follows:

[0111] ① Visitors entering the website → 2D digital human automatically greets them;

[0112] ② The dialogue stream is fed into the intent recognition model in real time;

[0113] ③ If P(intent = "Want to learn more") is greater than or equal to the set threshold and the visitor meets the stay / click conditions, then trigger strategy engine 300;

[0114] ④ A 3D exhibition hall guide card / button will pop up;

[0115] ⑤ After a visitor clicks, they enter an immersive space via the 3D showroom access gateway 400, where they are greeted by a 3D digital human, reusing the product interests already collected.

[0116] ⑥ Behavioral data is written back to the database for continuous model training.

[0117] It has the following advantages:

[0118] 1. Improved conversion rate: A / B testing showed that the conversion rate of potential visitors clicking on the 3D showroom increased by 35%, and the overall form lead generation rate increased by 18%.

[0119] 2. Controllable costs: Low-value traffic remains in the 2D stage, and 3D rendering resources are only available to high-potential customers, saving approximately 40% of server computing power.

[0120] 3. Smooth user experience: No forced redirects, precise triggering based on intent, and reduced interruptive push notifications.

[0121] 4. Model self-learning: The data loop continuously updates the accuracy of intent recognition, and the system becomes more and more accurate the more it is used.

[0122] Please see Figure 5 In another embodiment, a visitor guidance method for a digital exhibition hall system integrating 2D and 3D digital humans includes the following steps:

[0123] S1: Generate a 2D digital human to communicate with visitors through the 2D digital human front-end component, and output the visitor's dialogue text stream;

[0124] S2: The intent recognition module performs intent recognition on the dialogue text stream output by the 2D digital human front-end component and outputs intent labels;

[0125] S3: The trigger strategy engine determines whether to trigger 3D guidance based on the intent tags output by the intent recognition module. If it determines that 3D guidance is triggered, it guides the visitor into the 3D digital exhibition hall.

[0126] S4: 3D Exhibition Hall Access Gateway: Once it is confirmed that a user has entered the 3D digital exhibition hall, the visitor's identity, contextual intent, and product preferences are pushed to the 2D digital human front-end component.

[0127] S5: When the 2D digital human front-end component receives the packaged information pushed back from the 3D exhibition hall access gateway, it will jump to the 3D digital human tour guide system.

[0128] S6: The 3D digital human guide system generates personalized routes and narration scripts based on the user's visitor identity, contextual intent, and product preferences. It uses 3D digital humans to greet visitors and renders the 3D digital exhibition hall in real time.

[0129] When receiving visitors, the 2D digital human front-end component first generates a 2D digital human to communicate with the visitor, handling the initial dialogue, FAQs (Frequently Asked Questions), and information delivery. It then outputs the visitor's dialogue text stream to the intent recognition model. This model, trained on a large-scale labeled dialogue corpus, continuously monitors the dialogue text stream, outputs intent tags, and triggers a strategy engine. Based on the intent tags output by the intent recognition module, it determines whether to trigger 3D guidance. When 3D guidance is triggered, the 3D showroom access gateway pushes the visitor's identity, contextual intent, and product preferences to the 2D digital human front-end component. Upon receiving the packaged information pushed back from the 3D showroom access gateway, the 2D digital human front-end component redirects to the 3D showroom. The 3D digital human guide system generates personalized routes and narration scripts based on visitor identity, context, clothing, and product preferences. It guides visitors through a 3D digital human and renders a 3D digital exhibition hall in real time. Visitors are greeted by a 2D digital human, and the system automatically transitions them to the 3D digital human based on their intentions. This allows for precise switching from 2D chat to the 3D exhibition hall, avoiding a lack of immersion due to limited interaction with a flat chat interface, extending visitor dwell time, and keeping low-value traffic in the 2D environment. 3D rendering resources are only available to high-potential visitors, saving server computing power and thus reducing costs.

[0130] In some embodiments, the triggering strategy engine determines whether to trigger 3D guidance based on the intent label output by the intent recognition module, specifically including the following steps:

[0131] The trigger strategy engine determines whether to trigger 3D guidance by judging whether the intent tag is greater than the set threshold, the dwell time is greater than the preset time, and the keyword weight is greater than the preset weight.

[0132] The system determines whether 3D guidance needs to be triggered by judging whether the intent tag is greater than the set threshold, the dwell time is greater than the preset time, and the keyword weight. If the intent tag is greater than the set threshold, the dwell time is greater than the preset market, and the keyword weight is greater than the preset weight, it is determined that the visitor needs in-depth interaction, and 3D guidance is triggered to guide the visitor to the 3D exhibition hall.

[0133] In other embodiments, the determination of whether to trigger 3D guidance can also be made by judging whether the intent tag is greater than a set threshold, the dwell time is greater than a preset time, and the keyword weight, among one or more of these conditions.

[0134] In some embodiments, guiding visitors into the 3D digital exhibition hall specifically includes the following steps:

[0135] The trigger strategy engine guides visitors into the 3D digital exhibition hall by popping up a guide card or button.

[0136] Visitors are guided into the 3D digital exhibition hall by pop-up 3D digital exhibition hall guide cards or buttons. Customers can click on the 3D digital exhibition hall guide cards or buttons to confirm whether they want to enter the 3D digital exhibition hall.

[0137] In other embodiments, visitors can be asked by voice whether they need to enter the 3D digital exhibition hall.

[0138] In some embodiments, the step of packaging visitor identity, contextual intent, and product preferences and pushing them to the 2D digital human front-end component specifically includes the following steps:

[0139] The 3D showroom access gateway generates a one-time token and deep chain parameters, packages the visitor's identity, contextual intent, and product preferences, and pushes them to the 2D digital human front-end component in the form of a URL / QR-Code.

[0140] A One-Time Token (OTT) is a one-time password used for identity verification or authorization. It expires immediately after use and cannot be reused. Its extremely short validity period and automatic expiration upon expiration, coupled with an algorithm-generated unpredictable random value, prevent guessing or replay attacks and ensure visitor information is not leaked. Deep Link Parameters are key-value pairs embedded in deep links, used to pass specific data to the target application when a user clicks the link, enabling precise page redirection or personalized operations. A URL (Uniform Resource Locator) is an address standard used on the internet for locating and accessing resources; it is a common form of URI (Uniform Resource Identifier). A QR Code (Quick Response Code) is a matrix-style two-dimensional barcode capable of storing large amounts of data and being read quickly.

[0141] In some embodiments, the following steps are also included:

[0142] The data attribution and feedback module records the switching rate from 2D digital human to 3D digital human, visitor dwell time, and completed orders, and feeds this information back to the intent recognition model and trigger strategy engine.

[0143] By recording metrics such as the switching rate from 2D to 3D digital humans, visitor dwell time, and number of orders, the data is fed back to the intent recognition model and trigger strategy engine for reverse updates, making intent recognition more accurate and determining whether to provide 3D guidance more precise.

[0144] Finally, it should be noted that although the above embodiments have been described in the text and drawings of this application, this should not limit the scope of patent protection of this application. Any technical solutions that are based on the essential concept of this application and utilize the content described in the text and drawings of this application, resulting in equivalent structural or procedural substitutions or modifications, as well as the direct or indirect application of the technical solutions of the above embodiments to other related technical fields, are all included within the scope of patent protection of this application.

Claims

1. A digital exhibition hall system integrating 2D digital humans and 3D digital humans, characterized in that, include: The 2D digital human front-end component is used to communicate with visitors through a 2D digital human and output the visitor dialogue text stream. When it receives the packaged information pushed back from the 3D exhibition hall access gateway, it jumps to the 3D digital human tour guide system. The intent recognition module is used to recognize the intent of the dialogue text stream output by the 2D digital human front-end component and output intent tags. The trigger strategy engine is used to determine whether to trigger 3D guidance based on the intent label output by the intent recognition module. When it is determined that 3D guidance is triggered, the visitor is guided into the 3D digital exhibition hall. The 3D exhibition hall access gateway is used to package the visitor's identity, contextual intent, and product preferences and push them to the 2D digital human front-end component after determining that a user has entered the 3D digital exhibition hall. A 3D digital human tour guide system is used to generate personalized routes and narration scripts based on the visitor's identity, contextual intent, and product preferences. The system uses 3D digital humans to receive visitors and renders 3D digital exhibition halls in real time.

2. The digital exhibition hall system integrating 2D and 3D digital humans according to claim 1, characterized in that, The triggering strategy engine is specifically used to determine whether the intent tag is greater than a set threshold, the dwell time is greater than a preset time, and the keyword weight is greater than a preset weight. If so, it determines to trigger 3D guidance.

3. The digital exhibition hall system integrating 2D and 3D digital humans according to claim 1, characterized in that, The trigger strategy engine is specifically used to guide visitors into the 3D digital exhibition hall by popping up a 3D digital exhibition hall guide card or button.

4. The digital exhibition hall system integrating 2D and 3D digital humans according to claim 1, characterized in that, The 3D exhibition hall access gateway is used to package visitor identity, contextual intent, and product preferences into a URL / QR-Code format by generating a one-time token and deep chain parameters, and push it to the 2D digital human front-end component.

5. The digital exhibition hall system integrating 2D and 3D digital humans according to claim 1, characterized in that, Also includes: The data attribution and feedback module records the switching rate from 2D digital human to 3D digital human, visitor dwell time, and completed orders, and feeds this information back to the intent recognition model and trigger strategy engine.

6. A visitor guidance method for a digital exhibition hall system integrating 2D and 3D digital humans, characterized in that, Includes the following steps: The 2D digital human front-end component generates a 2D digital human to communicate with visitors and outputs the visitor's dialogue text stream. The intent recognition module performs intent recognition on the dialogue text stream output by the 2D digital human front-end component and outputs intent labels; The trigger strategy engine determines whether to trigger 3D guidance based on the intent tags output by the intent recognition module. If it determines that 3D guidance is triggered, it guides visitors into the 3D digital exhibition hall. Once the 3D exhibition hall access gateway confirms that a user has entered the 3D digital exhibition hall, it packages the visitor's identity, contextual intent, and product preferences and pushes them to the 2D digital human front-end component. When the 2D digital human front-end component receives the packaged information pushed back from the 3D exhibition hall access gateway, it will jump to the 3D digital human tour guide system; The 3D digital human guide system generates personalized routes and narration scripts based on the user's visitor identity, contextual intent, and product preferences. It uses 3D digital humans to greet visitors and renders 3D digital exhibition halls in real time.

7. The visitor guidance method for the digital exhibition hall system integrating 2D and 3D digital humans according to claim 6, characterized in that, The triggering strategy engine determines whether to trigger 3D guidance based on the intent tags output by the intent recognition module, specifically including the following steps: The trigger strategy engine determines whether to trigger 3D guidance by judging whether the intent tag is greater than the set threshold, the dwell time is greater than the preset time, and the keyword weight is greater than the preset weight.

8. The visitor guidance method for the digital exhibition hall system integrating 2D and 3D digital humans according to claim 6, characterized in that, Guiding visitors into the 3D digital exhibition hall specifically includes the following steps: The trigger strategy engine guides visitors into the 3D digital exhibition hall by popping up a guide card or button.

9. The visitor guidance method for the digital exhibition hall system integrating 2D and 3D digital humans according to claim 6, characterized in that, The process of packaging visitor identity, contextual intent, and product preferences and pushing them to the 2D digital human front-end component specifically includes the following steps: The 3D showroom access gateway generates a one-time token and deep chain parameters, packages the visitor's identity, contextual intent, and product preferences, and pushes them to the 2D digital human front-end component in the form of a URL / QR-Code.

10. The visitor guidance method for the digital exhibition hall system integrating 2D and 3D digital humans according to claim 6, characterized in that, It also includes the following steps: The data attribution and feedback module records the switching rate from 2D digital human to 3D digital human, visitor dwell time, and completed orders, and feeds this information back to the intent recognition model and trigger strategy engine.

Images