Scenarized interactive content optimization method and system based on virtual reality

Abstract

The invention discloses a scenarized interactive content optimization method and system based on virtual reality, and relates to the technical field related to data processing, and the method comprises the steps: constructing a virtual-real structure network platform, and carrying out the structural modeling of an offline scene, virtual content and user behavior; triggering loading of a virtual reality interaction scene through code scanning of a user, spatially sensing and verifying the real position of the user, and obtaining an activated virtual interaction instance after the verification is passed; starting a virtual scene, collecting user behavior data, and evaluating an interaction content effect from local and global two dimensions; and performing interactive content configuration optimization according to the integrated interactive content effect evaluation result. The technical problems that in the prior art, virtual reality interaction content is separated from an offline physical scene, and dynamic evaluation and closed-loop optimization capability based on multi-dimensional behavior data are lacked are solved, and accurate alignment, dynamic response and personalized interaction of virtual and real scenes are achieved. And the immersion, participation and satisfaction of the user in the virtual-real fusion environment are improved.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, specifically to a method and system for optimizing scene-based interactive content based on virtual reality. Background Technology

[0002] Traditional virtual reality (VR) interaction focuses on immersive visual experiences or pre-defined linear interaction processes, lacking the ability to dynamically respond to the user's physical environment and real-time behavior. This results in a disconnect between virtual content and the offline environment, making it difficult to achieve a natural, coherent, and personalized interactive experience. Meanwhile, the difficulty in scientifically evaluating the effectiveness of interactive content and achieving continuous optimization is a key constraint on the large-scale deployment and long-term operation of scenario-based VR applications. Existing VR interaction solutions mostly employ centralized content distribution and rendering mechanisms, with their interaction logic independent of the user's specific spatial location and environmental characteristics. Although some attempts are made to connect to offline scenes through lightweight entry points such as QR codes, this only achieves one-way content triggering and lacks unified modeling and real-time feedback on user state, environmental context, and multi-user collaborative behavior. Furthermore, the evaluation of virtual content effectiveness typically relies on subjective questionnaires or simple behavioral statistics, lacking a multi-dimensional, structured analytical framework from local interaction details to the overall user experience. This results in a lack of data support for optimization decisions and makes it difficult to achieve deep adaptation between interactive content and scene characteristics.

[0003] Therefore, current technologies suffer from several technical problems, including a disconnect between virtual reality interactive content and offline physical scenarios, and a lack of dynamic evaluation and closed-loop optimization capabilities based on multi-dimensional behavioral data. Summary of the Invention

[0004] This application provides a method and system for optimizing scenario-based interactive content based on virtual reality, which solves the technical problems in the prior art of the separation between virtual reality interactive content and offline physical scenes, and the lack of dynamic evaluation and closed-loop optimization capabilities based on multi-dimensional behavioral data. It achieves the technical effect of realizing precise alignment of virtual and real scenes, dynamic response and personalized interaction, and improving users' immersion, participation and satisfaction in the virtual and real integrated environment.

[0005] This application provides a method for optimizing scenario-based interactive content based on virtual reality. The method includes: constructing a virtual-real structure network platform for offline interactive scenarios, wherein the virtual-real structure network platform performs structured modeling of offline scenarios, virtual interactive content, and user behavior; triggering the loading of virtual reality interactive scenarios on the virtual-real structure network platform based on user scanning behavior, and verifying the user's real location through spatial perception; updating the interaction state in the virtual-real structure network platform when the verification is successful to obtain an activated virtual interaction instance; launching the virtual scene of the virtual-real structure network platform according to the activated virtual interaction instance; collecting user behavior data during virtual reality interaction, and performing a two-dimensional interactive content effect evaluation based on local and global features to obtain an integrated interactive content effect evaluation result; and optimizing the interactive content configuration of the virtual scene in the virtual-real structure network according to the integrated interactive content effect evaluation result to obtain an optimized virtual scene interactive content result.

[0006] In one possible implementation, a virtual-real structured network platform for offline interactive scenarios is constructed. This platform performs structured modeling of offline scenarios, virtual interactive content, and user behavior, including: creating offline scenarios on the management end to obtain a set of offline scenarios, where each offline scenario includes a scenario ID, scenario type, large-screen device ID, and a valid time interval; traversing the offline scenario set to configure virtual interactive content from three dimensions: game type, game parameters, and reward rules, to obtain an offline scenario-virtual interactive content configuration set; constructing user nodes, scenario nodes, game nodes, and reward nodes on the interaction end, and setting abstract rules for user behavior on these nodes to obtain a user behavior interaction edge generation mechanism; and constructing the virtual-real structured network platform based on the offline scenario-virtual interactive content configuration set, the user behavior interaction edge generation mechanism, and the user nodes, scenario nodes, game nodes, and reward nodes.

[0007] In one possible implementation, the virtual reality interactive scene loading of the virtual-real structure network platform is triggered based on the user's scanning behavior, including: when the target user scans the QR code, the virtual-real structure network platform parses the QR code parameters to obtain the scene ID and device ID for scene validity verification. If the verification is successful, the target user ID is obtained; the target user ID is bound to the virtual interactive content selected by the target user, and the virtual reality interactive scene is loaded through the virtual-real structure network platform.

[0008] In a possible implementation, the user's real-world location is verified through spatial perception. When the verification passes, the interaction state in the virtual-real structure network platform is updated to obtain an activated virtual interaction instance. This includes: collecting the target user's real-world location through a camera and infrared sensing device, and verifying the user's position against a preset interaction area. If the verification fails, the interaction is terminated. If the verification passes, the large screen provides virtual feedback that the animation has been activated, and the multi-source sensing device is activated and the interaction state in the virtual-real structure network platform is updated to obtain an activated virtual interaction instance.

[0009] In a possible implementation, the virtual scene of the virtual-real structure network platform is launched according to the activated virtual interaction instance. User behavior data is collected during the virtual reality interaction process, and a dual-dimensional evaluation of the interactive content effect, encompassing both local and global features, is performed to obtain an integrated interactive content effect evaluation result. This includes: collecting user behavior data from four dimensions—operation frequency, operation type, current score, and whether a reward has been redeemed—to obtain initial user behavior data; using the initial user behavior data as anchor points, performing a related data retrieval in the data repository of the virtual-real structure network platform to obtain a related user behavior data set; performing a global feature interactive content effect evaluation on the related user behavior data set to obtain global interactive content effect features; performing a local feature interactive content effect evaluation on the initial user behavior data to obtain local interactive content effect features; and integrating the global and local interactive content effect features to obtain the integrated interactive content effect evaluation result.

[0010] In a possible implementation, the global interactive content effect evaluation is performed on the associated user behavior data set to obtain the global interactive content effect features, including: traversing the associated user behavior data set to extract behavioral features and obtain an associated user behavior feature set; performing global feature mean shift filtering based on the associated user behavior feature set to determine the central associated user behavior features; and evaluating the interactive content effect based on the central associated user behavior features to obtain the global interactive content effect features.

[0011] In a possible implementation, integrating global interactive content effect features and local interactive content effect features to obtain the integrated interactive content effect evaluation result includes: calculating the feature element similarity between the global interactive content effect features and the local interactive content effect features to obtain a feature element similarity set; normalizing the feature element similarity set to construct an integrated interaction matrix; and using the integrated interaction matrix to enhance the interaction of the local interactive content effect features to obtain the integrated interactive content effect evaluation result.

[0012] In a possible implementation, the virtual scene in the virtual-real structure network is optimized for interactive content configuration based on the integrated interactive content effect evaluation result to obtain the virtual scene interactive content optimization result. This includes: acquiring the target interactive content effect; identifying deviations from the integrated interactive content effect evaluation result to obtain the deviation direction and degree; optimizing the virtual scene for directional interactive content configuration based on the deviation direction and degree to obtain the initial virtual scene interactive content optimization result; calling the data repository of the virtual-real structure network platform to predict the optimization effect of the initial virtual scene interactive content optimization result; and determining whether the target interactive content effect is met based on the prediction result. If so, the initial virtual scene interactive content optimization result is used as the virtual scene interactive content optimization result.

[0013] In a possible implementation, the virtual reality-based scenario-based interactive content optimization method further includes: if not, optimizing the initial virtual scene interactive content optimization result again based on the target interactive content effect until the prediction result satisfies the target interactive content effect.

[0014] This application also provides a virtual reality-based scenario-based interactive content optimization system, comprising: a platform construction module for constructing a virtual-real structure network platform for offline interactive scenarios, wherein the virtual-real structure network platform performs structured modeling of offline scenarios, virtual interactive content, and user behavior; an interaction instance acquisition module for triggering the loading of virtual reality interactive scenarios on the virtual-real structure network platform based on user scanning behavior, verifying the user's real location through spatial perception, and updating the interaction state in the virtual-real structure network platform when the verification is successful, thereby obtaining an activated virtual interactive instance; an evaluation result acquisition module for launching the virtual scene of the virtual-real structure network platform according to the activated virtual interactive instance, collecting user behavior data during virtual reality interaction, and performing a two-dimensional evaluation of the interactive content effect based on local and global features, thereby obtaining an integrated interactive content effect evaluation result; and an optimization result acquisition module for optimizing the interactive content configuration of the virtual scene in the virtual-real structure network according to the integrated interactive content effect evaluation result, thereby obtaining an optimized virtual scene interactive content result.

[0015] This application proposes a virtual reality-based scenario-based interactive content optimization method and system. It constructs a virtual-real structure network platform to structurally model offline scenes, virtual content, and user behavior. The system triggers the loading of a virtual reality interactive scene by user scanning a QR code, verifies the user's real-world location using spatial perception, and activates the virtual interaction instance upon successful verification. The system then initiates the virtual scene and collects user behavior data, evaluating the interactive content's effectiveness from both local and global dimensions. Based on the integrated evaluation results, the system optimizes the interactive content configuration. This addresses the technical problems of existing technologies, such as the disconnect between virtual reality interactive content and offline physical scenes, and the lack of dynamic evaluation and closed-loop optimization capabilities based on multi-dimensional behavioral data. It achieves precise alignment of virtual and real scenes, dynamic response, and personalized interaction, enhancing users' immersion, engagement, and satisfaction in the virtual-real integrated environment. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of this disclosure, the accompanying drawings of the embodiments of this disclosure will be briefly described below. Flowcharts are used in this application to illustrate the operations performed by the system according to the embodiments of this application. It should be understood that the preceding or following operations are not necessarily performed precisely in sequence. Instead, various steps can be processed in reverse order or simultaneously as needed. Furthermore, other operations can be added to these processes, or one or more steps can be removed from these processes.

[0017] Figure 1 This is a schematic diagram of the process for optimizing scene-based interactive content based on virtual reality, as provided in an embodiment of this application.

[0018] Figure 2 This is a schematic diagram of the structure of a virtual reality-based scenario-based interactive content optimization system provided in an embodiment of this application.

[0019] Figure labeling: Platform construction module 10, interactive instance acquisition module 20, evaluation result acquisition module 30, optimization result acquisition module 40. Detailed Implementation

[0020] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description is provided in conjunction with the accompanying drawings and preferred embodiments, based on the specific implementation methods, structure, features, and effects of the present invention.

[0021] This application provides a method for optimizing contextualized interactive content based on virtual reality, such as... Figure 1 As shown, the method includes: Step S100: Construct a virtual-real structure network platform for offline interactive scenarios, wherein the virtual-real structure network platform performs structured modeling of offline scenarios, virtual interactive content, and user behavior.

[0022] Step S100 further includes: creating offline scenarios on the management end to obtain a set of offline scenarios, wherein each offline scenario includes a scenario ID, scenario type, large screen device ID, and a valid time interval for the scenario; traversing the set of offline scenarios to configure virtual interactive content from three dimensions: game type, game parameters, and reward rules, to obtain an offline scenario-virtual interactive content configuration set; constructing user nodes, scenario nodes, game nodes, and reward nodes on the interaction end, and setting user behavior abstraction rules for the user nodes, scenario nodes, game nodes, and reward nodes to obtain a user behavior interaction edge generation mechanism; and constructing the virtual-real structure network platform based on the offline scenario-virtual interactive content configuration set, the user behavior interaction edge generation mechanism, and the user nodes, scenario nodes, game nodes, and reward nodes.

[0023] Preferably, digital location information is established for each offline interactive location in the backend management interface, resulting in a set of offline scenarios. Each offline scenario includes a scenario ID, scenario type, large screen device ID, and scenario validity time interval. Specifically, the scenario ID is a unique identifier used to accurately locate and call the physical location in the system; the scenario type is a classification of the location's purpose, such as a shopping mall atrium advertising screen, a museum exhibition area entrance, or a station information kiosk; the large screen device ID is a unique identifier for a specific display screen bound to the physical location and used to display virtual interactive content such as VR game screens, used to establish a hard association between the venue and the output device; the scenario validity time interval is used to define the effective operating period of the interactive location, enabling control over the activation and deactivation of interactive functions.

[0024] Preferably, each physical location in the offline scene set is traversed, and its corresponding virtual interactive content is configured from three dimensions: game type, game parameters, and reward rules. Game type is used to select the type of interactive program, such as first-person exploration, motion-sensing fruit cutting, or quiz challenges. Game parameters are used to set specific variables for the selected interactive type, such as game duration, difficulty level, target score, and virtual item attributes. Reward rules are used to define the incentives that users can obtain after achieving their goals, such as redeeming a discount coupon, earning points, unlocking a digital badge, and the conditions and methods for redeeming the reward. This results in an offline scene-virtual interactive content configuration set, containing multiple sets of mapping relationships between offline locations and virtual interactive content, and clearly recording the interactive experience of each physical location with different types and dimensions of content.

[0025] Preferably, user nodes, scene nodes, game nodes, and reward nodes are constructed on the interaction end as nodes in the data graph. User nodes represent individuals participating in the interaction, scene nodes represent created physical locations, game nodes represent configured interactive program instances, and reward nodes represent configured reward item instances. User behavior abstraction rules are set for user nodes, scene nodes, game nodes, and reward nodes, specifying the behavior of users establishing relationships between any two types of nodes. This results in a user behavior interaction edge generation mechanism. For example, when user A starts game C in scene B, a participation edge from user A node to game C node and a membership edge from game C node to scene B node are automatically created. Finally, the offline scene-virtual interactive content configuration set, the user behavior interaction edge generation mechanism, and the four types of nodes (user, scene, game, and reward) are integrated into the data platform to determine the virtual-real structure network platform. This platform is used for central data processing and scheduling, enabling the invocation of corresponding game content and devices based on scene IDs, dynamic generation and updating of the graph network in response to user behavior, and recording of complete interaction paths. It also provides a structured data foundation for effect evaluation and optimization.

[0026] Step S200: Based on the user's scanning behavior, the virtual reality interactive scene of the virtual-real structure network platform is loaded, and the user's real position is verified by spatial perception. When the verification is successful, the interaction state in the virtual-real structure network platform is updated to obtain an activated virtual interaction instance.

[0027] Step S200 further includes: after the target user scans the QR code, the virtual-real structure network platform parses the QR code parameters to obtain the scene ID and device ID for scene validity verification. If the verification is successful, the target user ID is obtained; the target user ID is bound to the virtual interactive content selected by the target user, and the virtual reality interactive scene is loaded through the virtual-real structure network platform.

[0028] Preferably, based on the user's scanning behavior, the virtual reality interactive scene loading of the virtual-physical structure network platform is triggered. Specifically, when the target user scans the QR code deployed in the offline scene using their smart terminal, the user's scanning action sends a request to the virtual-physical structure network platform. After receiving the request, the platform parses the QR code parameters, including at least obtaining the scene ID and device ID. The scene ID is used to uniquely identify the physical location where the user is located or is pointing, and the device ID is used to uniquely identify the target output device expected to display VR content after the interaction is triggered, such as a specific large screen. Then, the parsed scene ID and device ID are used to set up the offline scene. The scene is validly verified through cameras and infrared sensing devices. The verification may include existence verification to confirm whether the scene ID and device ID are registered and valid; association verification to confirm whether the device ID is indeed bound to the scene ID; and status verification to determine whether the interaction is allowed to be activated at the current moment based on attributes such as the valid time interval of the scene. If all scene valid verifications pass, the operation of obtaining the target user's identity is performed, such as guiding the user to authorize login and obtaining their unique user ID within the ecosystem.

[0029] Preferably, based on the verified scene ID, the system retrieves the preset virtual interactive content options or default content for the scene from the offline scene-virtual interactive content configuration set, creates a logical interaction instance in the background, and binds the obtained target user ID with the retrieved virtual interactive content. That is, all operations, scores, results, and other data of this interactive experience are associated with this user. Then, the virtual-real structure network platform sends loading instructions and necessary scene data packets to the specified large screen or the user's VR headset or other terminal device according to the bound content and device ID. After receiving the corresponding instructions, the target device starts the corresponding rendering engine, loads and presents the bound virtual reality interactive scene, such as a 3D game environment or a 360° panoramic video interface. At the same time, in the background of the virtual-real structure network platform, the system updates the status of the interaction instance and creates initial edges in the structure network graph according to the user behavior interaction edge generation mechanism, such as connecting user nodes and game nodes.

[0030] Furthermore, step S200 also includes collecting the target user's real-world location using a camera and infrared sensing device, and verifying the location against a preset interactive area. If the verification fails, the interaction is terminated; if the verification passes, the large screen provides virtual feedback that the animation has been activated, and the interactive multi-source sensing device is activated and the interactive state in the virtual-real structure network platform is updated to obtain an activated virtual interactive instance.

[0031] Preferably, the system collects the target user's real-world location by capturing video streams of the area where the target user is located via a camera, and using object detection and human skeletal key point recognition to analyze the images and estimate the user's position on a two-dimensional plane (X, Y axes). Infrared sensing devices such as Kinect or infrared structured light-based depth cameras emit and receive reflected infrared light patterns to accurately measure the distance from the user to the device, i.e., the Z-axis depth. The planar position information acquired by the camera is then fused with the depth information acquired by the infrared device to calculate the target user's precise real-world coordinates in a three-dimensional spatial coordinate system. The preset interactive area refers to one or more three-dimensional spatial ranges defined in the system backend for this offline scene, such as a rectangular space 1.5 to 3 meters in front of the large screen, with a width of 2 meters, serving as a safe / effective interactive zone. The system verifies the target user's precise real-world location coordinates against the boundaries of the preset interaction area. If the user's coordinates are outside the preset interaction area, the verification fails, the core interaction logic is not initiated, and a message "Please move forward to the designated area" may be displayed on the large screen, terminating the current interaction process. The user is then left to reposition or abandon the interaction. This prevents users from being unable to experience the interaction properly due to being too close or too far away, or prevents non-target users from accidentally triggering the interaction, thus ensuring the safety and effectiveness of the interactive experience. If the verification passes, the system controls the large screen's virtual feedback animation to activate, such as displaying dynamic effects like "Welcome" or "Game loading," giving the user a clear and positive visual signal that the interaction has begun. Simultaneously, the system switches to high-precision mode and activates the function to capture the user's location. The interactive multi-source sensing device includes at least a high frame rate camera for more precise gesture recognition, an inertial measurement unit, and a microphone array to capture voice commands. Simultaneously, it updates the interactive state in the virtual-real network platform, changing the state attribute of this interactive session from "pending activation" or "verification in progress" to "activated" or "in progress." Ultimately, the instantiation operations such as user arrival, hardware readiness, and state updates logically constitute a complete and executable activated virtual interactive instance. This is a dynamic combination and operating carrier of offline scene nodes, user nodes, game content nodes, and the current state of all hardware devices in a specific time and space, thereby ensuring that virtual interactive content can unfold safely and smoothly based on correct physical space information and user state.

[0032] Step S300: Activate the virtual scene of the virtual-real structure network platform according to the activated virtual interaction instance, collect user behavior data during the virtual reality interaction process, and conduct a two-dimensional interactive content effect evaluation based on local and global features to obtain an integrated interactive content effect evaluation result.

[0033] Step S300 further includes collecting user behavior data from four dimensions: operation frequency, operation type, current score, and whether the reward has been redeemed, to obtain initial user behavior data; using the initial user behavior data as anchor points, performing related data retrieval in the data repository of the virtual-physical network platform to obtain a set of related user behavior data; performing a global feature interactive content effect evaluation on the set of related user behavior data to obtain global interactive content effect features; performing a local feature interactive content effect evaluation on the initial user behavior data to obtain local interactive content effect features; and integrating the global interactive content effect features and the local interactive content effect features to obtain the integrated interactive content effect evaluation result.

[0034] Preferably, a startup command is sent to rendering terminals such as large screens and VR devices upon activation of the virtual interactive instance, initiating the virtual scene of the virtual-real structure network platform, officially running and displaying the configured virtual reality interactive content, such as games or simulation programs. Then, during the operation of the interactive program, the original behavior log of this interactive instance is recorded in real time through embedded points or event listening. User behavior data is collected from four preset dimensions: operation frequency, operation type, current score, and whether rewards have been redeemed. Among them, operation frequency refers to the number of times the user triggers interactive events such as clicking, waving, and jumping per unit time, reflecting the user's participation activity; operation type refers to the specific interactive action category performed by the user, such as moving, selecting, attacking, and asking for help, reflecting the user's strategy preference or operation mode; current score refers to the user's progress or achievement value calculated and fed back by the interactive program in real time according to built-in rules, reflecting the user's immediate performance and achievement; whether rewards have been redeemed records whether the user has actually redeemed preset rewards such as scanning codes to receive coupons or crediting points during or after the interaction; thus determining the initial user behavior data.

[0035] Preferably, the initial user behavior data is used as the anchor point. Key fields such as user ID, scene ID, game ID, and timestamp in the initial user behavior data are used as query conditions to perform related data retrieval in the data repository of the virtual and physical structure network platform. This yields historical behavior data of the same user, including the user's interaction records at other times and in other scenarios; behavior data of other users in the same scenario, including the interaction history of other users at that physical location; and behavior data of other users under the same game content, including records of all users who have experienced the interactive content. This allows for the determination of a set of related user behavior data associated with the current interaction instance in terms of user, scene, and content dimensions.

[0036] Preferably, statistical analysis and cluster analysis are used to evaluate the global characteristics of interactive content effectiveness on the associated user behavior data set. This involves analyzing from a macro perspective of group behavior and long-term trends to extract stable characteristics that are not dependent on a single session. Examples include the average participation time of the game content, the general difficulty curve of users, the common operation patterns of high-conversion user groups, and the changing patterns of participation popularity over different time periods. These global interactive content effectiveness characteristics reflect the comprehensive effect of the interactive content across a wide user group and a long time span. Next, local interactive content effectiveness evaluation is performed on the initial user behavior data. This involves direct real-time calculation and analysis to capture the micro-details and real-time state characteristics of the current specific interaction. Examples include whether the current operation sequence is abnormal, whether the score growth rate meets expectations, and whether a reward redemption action has occurred in the current stage. This yields local interactive content effectiveness characteristics that reflect the real-time effect of the interactive content in that specific real-time interactive session. By calculating feature similarity, constructing an interaction matrix, and performing feature enhancement, the global interactive content effect features and local interactive content effect features are integrated. For example, the global feature "most users spend an average of 2 minutes on this level" is used as a benchmark to evaluate whether the local feature "this user passed in 30 seconds" is exceptionally good or may exploit a vulnerability, ultimately generating a more comprehensive and accurate integrated interactive content effect evaluation result.

[0037] Furthermore, step S300 also includes: traversing the associated user behavior data set to extract behavioral features and obtain an associated user behavior feature set; performing global feature mean shift filtering based on the associated user behavior feature set to determine the central associated user behavior features; and evaluating the interactive content effect based on the central associated user behavior features to obtain global interactive content effect features.

[0038] Preferably, the associated user behavior data set is traversed to extract behavioral features. For each record, more representative and analyzable numerical features are calculated from the original fields. For example, the operation type sequence is transformed into statistical features such as "attack operation ratio" and "movement operation variance". The operation frequency and participation duration are combined to calculate the "average operation interval". "Score growth rate" and "checkpoint dwell time" are extracted from the score sequence. Whether the transaction is cancelled is used as a binary feature, thereby obtaining an associated user behavior feature set containing multiple feature vectors. Then, a global feature mean shift filter is performed on the associated user behavior feature set. That is, for each point in the space, iteratively moves towards the direction with higher point density in its neighborhood until it converges to a local peak of the density distribution. Specifically, a kernel function and bandwidth are set for each feature vector, the mean of all points within the bandwidth of that point is calculated, and the point is moved to this mean position. This process is repeated until the position of the point no longer changes significantly. All points that converge to the same position are considered to belong to the same pattern, thus effectively filtering out discrete outliers, such as behavior, misoperation, data from extremely skilled or novice users, and determining the region with the densest data distribution that best represents the behavior pattern of the mainstream user group. Finally, the convergence point with the densest data points is taken as the central associated user behavior feature, that is, the feature vector calculated that best represents the typical or common user behavior pattern in the historical associated dataset.

[0039] Preferably, the user behavior characteristics associated with the central feature are used as a benchmark for evaluating the effectiveness of interactive content. This includes evaluating the effectiveness of interactive content from multiple dimensions. For example, the "average operation frequency" and "participation duration" in the central feature reflect the average depth of user participation, while the "average operation interval" and "error rate" reflect the smoothness of the operation design. The distribution of the "final score" or "redemption rate" in the central feature reflects the balance between the challenge of the content and the attractiveness of the reward. By analyzing the speed at which user behavior characteristics converge to the central feature at different stages, the learning cost of the content is evaluated, and finally, the overall interactive content effectiveness characteristics are obtained. For example, the content attracts users to engage for an average of 5 minutes, and the mainstream operation mode is smooth, but the final reward redemption conversion rate is only 15%, indicating that the matching degree of the attractiveness of the challenge and the reward needs to be optimized. This truly reflects the universal applicability and long-term attractiveness of the interactive content to the target user group.

[0040] Furthermore, step S300 also includes calculating the feature element similarity between the global interactive content effect features and the local interactive content effect features to obtain a feature element similarity set; normalizing the feature element similarity set to construct an integrated interaction matrix; and using the integrated interaction matrix to enhance the interaction of the local interactive content effect features to obtain an integrated interactive content effect evaluation result.

[0041] Preferably, a similarity measurement function is selected according to the type of feature element, and the similarity of feature elements of global interactive content effect features and local interactive content effect features is calculated pairwise. Specifically, for numerical features such as frequency and score, cosine similarity, the inverse of Euclidean distance, or normalized difference is commonly used. For categorical / state features, such as whether or not a feature is cancelled, Jaccard similarity can be used or the equality can be directly determined. Then, the set of feature element similarities is calculated and determined, where each element is a value between 0 and 1, representing the degree of matching or deviation of the local feature from the global benchmark in a certain dimension. Then, the similarity set of feature elements is normalized using Min-Max normalization or Z-Score normalization. Based on interaction design experience, the weighted influence relationships between different feature dimensions are predefined, or a correlation or causal relationship matrix between different feature dimensions is learned through analysis of a large amount of historical session data. An integrated interaction matrix is ​​then created to characterize the interactions between different feature dimensions. This matrix defines the corrective effect of the local-global similarity of one dimension on the local evaluation result of another dimension. Matrix elements represent the global-local similarity of the j-th feature dimension, and the weight or influence factor when enhancing or correcting the evaluation result of the i-th local feature dimension. The integrated interaction matrix is ​​used to enhance the interaction features of local interactive content. This involves performing matrix transformations or attention mechanisms, calculating the difference vector between global and local features, and then multiplying this difference vector by the integrated interaction matrix for component weighting. For indicators with high local-global similarity, the difference between the global pattern and the current pattern is introduced to a greater extent for correction; for indicators with significant differences, the correction of the global pattern is weakened or ignored, preserving more of the original information of the local data. The final evaluation results of the integrated interactive content include not only the specific details and real-time status of this interactive instance, but also the statistical stability and trend information of historical macro data to robustly enhance local indicators that conform to the norm, while prudently retaining indicators that deviate significantly, thereby improving the robustness and accuracy of the interactive effect evaluation.

[0042] Step S400: Based on the evaluation results of the integrated interactive content effect, optimize the interactive content configuration of the virtual scene in the virtual-real structure network to obtain the virtual scene interactive content optimization results.

[0043] Step S400 further includes: acquiring the target interactive content effect; identifying deviations in the integrated interactive content effect evaluation results to obtain the deviation direction and degree; optimizing the directional interactive content configuration of the virtual scene based on the deviation direction and degree to obtain the initial virtual scene interactive content optimization result; calling the data repository of the virtual-real structure network platform to predict the optimization effect of the initial virtual scene interactive content optimization result; and determining whether the target interactive content effect is met based on the prediction result. If so, the initial virtual scene interactive content optimization result is used as the virtual scene interactive content optimization result.

[0044] Furthermore, step S400 also includes, if not, optimizing the initial virtual scene interactive content optimization result again based on the target interactive content effect until the prediction result meets the target interactive content effect.

[0045] Preferably, the interactive content configuration in the virtual scene of the virtual-real structure network is optimized by using the integrated interactive content effect evaluation results. Specifically, the target interactive content effect, i.e., the expected effect indicators, such as average participation time ≥ 5 minutes, redemption conversion rate ≥ 30%, and user satisfaction score ≥ 4.5, is obtained. The deviation between these and the integrated interactive content effect evaluation results is then identified to obtain the direction and degree of deviation. The deviation direction indicates whether each effect indicator is not met or exceeds the target. Not meeting the target is a negative deviation, i.e., the redemption rate of the integrated interactive content effect evaluation results is less than the redemption rate of the target interactive content effect. Exceeding the target is a positive deviation, which may be that the target interactive content effect indicator is set too low. The degree of deviation is used to quantify the difference between each interactive content effect evaluation indicator. Then, configure the mapping rule library for performance metrics and content parameters. For example, if the redemption rate deviates negatively and the deviation is large, the optimization direction is to "reduce game difficulty" or "increase reward attractiveness"; if the "participation time" deviates negatively, the optimization direction is to "add level content" or "introduce narrative elements"; if the "operation error rate" deviates positively, the optimization direction is to "simplify the operation interface" or "enhance the new player guidance". Then, using the deviation direction and degree as input, match relevant optimization rules from the mapping rule library and convert them into adjustment instructions to optimize the directional interactive content configuration of the virtual scene. This includes at least game parameters such as difficulty coefficient, time limit, and life value, reward rules such as reward threshold and reward type, as well as the display format and delivery time of the interactive content, to determine the initial virtual scene interactive content optimization results.

[0046] Preferably, historical interaction data stored in the data repository of the virtual-real network platform is used to predict the optimization effect of the initial virtual scene interaction content optimization result. That is, the user behavior data expected to be generated and the effect evaluation index obtained by simulating the configuration parameters to be adjusted are used as the prediction result. Then, it is compared with the effect of the target interaction content. If the prediction result meets the effect of the target interaction content, the initial virtual scene interaction content optimization result is determined to be theoretically effective. The initial virtual scene interaction content optimization result is used as the virtual scene interaction content optimization result and finally updated to the offline scene-virtual interaction content configuration set. This completes the real-time or timed adjustment of the online interaction content and ensures that the prediction effect is consistent with the business and experience goals. If the predicted result does not meet the target interactive content effect, the initial virtual scene interactive content optimization result is optimized again based on the target interactive content effect. That is, the direction and degree of deviation between the predicted result and the target interactive content effect are calculated. Based on the new deviation information, the optimization strategy is adjusted, such as increasing the optimization intensity, adding optimization dimensions, or switching optimization rules, to generate a new virtual scene interactive content optimization result. The effect is then predicted again to obtain a new prediction result, which is compared with the target interactive content effect until it is met. Finally, the approved virtual scene interactive content optimization result is determined, and the necessary effectiveness of the optimization result is ensured, thereby improving the user's immersion, participation, and satisfaction in the virtual-real integrated environment.

[0047] In the above text, refer to Figure 1 A method for optimizing scene-based interactive content based on virtual reality, according to embodiments of the present invention, is described in detail. Next, reference will be made to... Figure 2 A virtual reality-based scenario-based interactive content optimization system is described according to an embodiment of the present invention.

[0048] The virtual reality-based scenario-based interactive content optimization system according to embodiments of the present invention addresses the technical problems in existing technologies, such as the disconnect between virtual reality interactive content and offline physical scenes, and the lack of dynamic evaluation and closed-loop optimization capabilities based on multi-dimensional behavioral data. It achieves precise alignment of virtual and real scenes, dynamic response, and personalized interaction, thereby enhancing users' immersion, participation, and satisfaction in a virtual-real integrated environment. Figure 2 As shown, the virtual reality-based scenario-based interactive content optimization system includes: a platform construction module 10, an interactive instance acquisition module 20, an evaluation result acquisition module 30, and an optimization result acquisition module 40.

[0049] Platform construction module 10 is used to construct a virtual-real structure network platform for offline interactive scenarios, wherein the virtual-real structure network platform performs structured modeling of offline scenarios, virtual interactive content, and user behavior; Interaction instance acquisition module 20 is used to trigger the loading of virtual reality interactive scenarios on the virtual-real structure network platform based on user scanning behavior, and verify the user's real location through spatial perception. When the verification is successful, the interaction state in the virtual-real structure network platform is updated to obtain an activated virtual interactive instance; Evaluation result acquisition module 30 is used to start the virtual scene of the virtual-real structure network platform according to the activated virtual interactive instance, collect user behavior data during virtual reality interaction, and perform dual-dimensional interactive content effect evaluation of local features and global features to obtain an integrated interactive content effect evaluation result; Optimization result acquisition module 40 is used to optimize the interactive content configuration of the virtual scene in the virtual-real structure network according to the integrated interactive content effect evaluation result to obtain virtual scene interactive content optimization result.

[0050] The specific configuration of platform construction module 10 will be described in detail below. Platform construction module 10 further includes: creating offline scenarios on the management end to obtain a set of offline scenarios, wherein each offline scenario includes a scenario ID, scenario type, large screen device ID, and scenario valid time interval; traversing the set of offline scenarios to configure virtual interactive content from three dimensions: game type, game parameters, and reward rules, to obtain an offline scenario-virtual interactive content configuration set; constructing user nodes, scenario nodes, game nodes, and reward nodes on the interaction end, and setting user behavior abstraction rules for the user nodes, scenario nodes, game nodes, and reward nodes to obtain a user behavior interaction edge generation mechanism; and constructing the virtual-real structure network platform based on the offline scenario-virtual interactive content configuration set, the user behavior interaction edge generation mechanism, and the user nodes, scenario nodes, game nodes, and reward nodes.

[0051] The specific configuration of the interactive instance acquisition module 20 will be described in detail below. The interactive instance acquisition module 20 further includes: when the target user scans the QR code, the virtual-real structure network platform parses the QR code parameters to obtain the scene ID and device ID for scene validity verification. If the verification is successful, the target user ID is obtained; the target user ID is bound to the virtual interactive content selected by the target user, and the virtual reality interactive scene is loaded through the virtual-real structure network platform.

[0052] The specific configuration of the interaction instance acquisition module 20 will be described in detail below. The interaction instance acquisition module 20 further includes: collecting the real location of the target user through a camera and infrared sensing device, and performing a position verification with a preset interaction area. If the verification fails, the interaction is terminated; if the verification passes, the large screen provides virtual feedback that the animation has been activated, and the multi-source sensing device for interaction is activated and the interaction state in the virtual-real structure network platform is updated to obtain an activated virtual interaction instance.

[0053] The specific configuration of the evaluation result acquisition module 30 will be described in detail below. The evaluation result acquisition module 30 further includes: collecting user behavior data from four dimensions—operation frequency, operation type, current score, and whether a reward has been redeemed—to obtain initial user behavior data; using the initial user behavior data as an anchor point, performing a related data retrieval in the data repository of the virtual-physical network platform to obtain a related user behavior data set; performing a global feature interactive content effect evaluation on the related user behavior data set to obtain global interactive content effect features; performing a local feature interactive content effect evaluation on the initial user behavior data to obtain local interactive content effect features; and integrating the global and local interactive content effect features to obtain the integrated interactive content effect evaluation result.

[0054] The specific configuration of the evaluation result acquisition module 30 will be described in detail below. The evaluation result acquisition module 30 further includes: traversing the associated user behavior data set to extract behavioral features and obtain an associated user behavior feature set; performing global feature mean shift filtering based on the associated user behavior feature set to determine the central associated user behavior features; and evaluating the interactive content effect based on the central associated user behavior features to obtain global interactive content effect features.

[0055] The specific configuration of the evaluation result acquisition module 30 will be described in detail below. The evaluation result acquisition module 30 further includes: calculating the feature element similarity between the global interactive content effect features and the local interactive content effect features to obtain a feature element similarity set; normalizing the feature element similarity set to construct an integrated interaction matrix; and using the integrated interaction matrix to enhance the interaction of the local interactive content effect features to obtain an integrated interactive content effect evaluation result.

[0056] The specific configuration of the optimization result acquisition module 40 will be described in detail below. The optimization result acquisition module 40 further includes: acquiring the target interactive content effect; identifying deviations in the integrated interactive content effect evaluation results to obtain the deviation direction and degree; optimizing the directional interactive content configuration of the virtual scene based on the deviation direction and degree to obtain an initial virtual scene interactive content optimization result; calling the data repository of the virtual-real structure network platform to predict the optimization effect of the initial virtual scene interactive content optimization result; and, based on the prediction result, determining whether the target interactive content effect is met. If so, the initial virtual scene interactive content optimization result is used as the virtual scene interactive content optimization result.

[0057] The specific configuration of the optimization result acquisition module 40 will be described in detail below. The optimization result acquisition module 40 further includes: if not, optimizing the initial virtual scene interaction content optimization result again based on the target interaction content effect until the prediction result meets the target interaction content effect.

[0058] The virtual reality-based scenario-based interactive content optimization system provided in this embodiment of the invention can execute the virtual reality-based scenario-based interactive content optimization method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0059] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any modifications, equivalent changes, and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A method for optimizing scenario-based interactive content based on virtual reality, characterized in that, The method includes: A virtual-real structured network platform for offline interactive scenarios is constructed, wherein the virtual-real structured network platform performs structured modeling of offline scenarios, virtual interactive content, and user behavior; The virtual reality interactive scene is loaded on the virtual and real structure network platform based on the user's scanning behavior, and the user's real position is verified by spatial perception. When the verification is successful, the interaction state in the virtual and real structure network platform is updated to obtain an activated virtual interaction instance. The virtual scene of the virtual-real structure network platform is launched according to the activated virtual interaction instance. User behavior data is collected during the virtual reality interaction process, and the effect of interactive content is evaluated from two dimensions: local features and global features, to obtain the integrated interactive content effect evaluation result. Based on the evaluation results of the integrated interactive content, the interactive content configuration of the virtual scene in the virtual-real structure network is optimized to obtain the virtual scene interactive content optimization results.

2. The method for optimizing scene-based interactive content based on virtual reality as described in claim 1, characterized in that, A virtual-real structured network platform for constructing offline interactive scenarios is established, wherein the virtual-real structured network platform performs structured modeling of offline scenarios, virtual interactive content, and user behavior, including: Create offline scenarios on the management side to obtain a set of offline scenarios. Each offline scenario includes a scenario ID, scenario type, large screen device ID, and scenario validity time range. The virtual interactive content configuration is configured by traversing the offline scene set from three dimensions: game type, game parameters, and reward rules, to obtain the offline scene-virtual interactive content configuration set. User nodes, scene nodes, game nodes, and reward nodes are constructed on the interactive end, and user behavior abstraction rules are set for the user nodes, scene nodes, game nodes, and reward nodes to obtain a user behavior interaction edge generation mechanism. Based on the offline scenario-virtual interactive content configuration set, the user behavior interaction edge generation mechanism, and the user node, scene node, game node, and reward node, the virtual-real structure network platform is constructed.

3. The method for optimizing scene-based interactive content based on virtual reality as described in claim 1, characterized in that, The loading of a virtual reality interactive scene on the virtual-real network platform is triggered based on the user's QR code scanning behavior, including: When a target user scans the QR code, the virtual-physical network platform parses the QR code parameters to obtain the scene ID and device ID for scene validity verification. If the verification is successful, the target user ID is obtained. The target user ID is bound to the virtual interactive content selected by the target user, and the virtual reality interactive scene is loaded through the virtual-real structure network platform.

4. The method for optimizing scene-based interactive content based on virtual reality as described in claim 1, characterized in that, And by verifying the user's real-world location through spatial awareness, when the verification passes, the interaction state in the virtual-real structure network platform is updated to obtain an activated virtual interaction instance, including: The system uses cameras and infrared sensors to collect the target user's real-world location and verify their position against a preset interactive area. If the verification fails, the interaction is terminated. If the verification passes, the large screen will provide virtual feedback that the animation has been activated, and the interactive multi-source sensing devices will be activated and the interactive state in the virtual-real structure network platform will be updated to obtain an activated virtual interactive instance.

5. The method for optimizing scene-based interactive content based on virtual reality as described in claim 1, characterized in that, The virtual scene of the virtual-real structure network platform is launched according to the activated virtual interaction instance. User behavior data is collected during the virtual reality interaction process, and the interactive content effect is evaluated from two dimensions: local features and global features. The integrated interactive content effect evaluation result is obtained, including: Initial user behavior data is obtained by collecting user behavior data from four dimensions: operation frequency, operation type, current score, and whether the reward has been redeemed. Using the initial user behavior data as an anchor point, related data retrieval is performed in the data repository of the virtual-real structure network platform to obtain a set of related user behavior data. A global feature-based interactive content effect evaluation is performed on the associated user behavior data set to obtain global interactive content effect features; The initial user behavior data is used to evaluate the effect of local feature interaction content to obtain local interaction content effect features; By integrating the global interactive content effect features and the local interactive content effect features, the integrated interactive content effect evaluation result is obtained.

6. The method for optimizing scene-based interactive content based on virtual reality as described in claim 5, characterized in that, A global feature-based interactive content effectiveness evaluation is performed on the associated user behavior data set to obtain the global interactive content effectiveness features, including: Traverse the associated user behavior data set to extract behavioral features and obtain the associated user behavior feature set. Based on the set of associated user behavior features, global feature mean shift filtering is performed to determine the central associated user behavior features; The effectiveness of interactive content is evaluated based on the user behavior characteristics associated with the central link, thereby obtaining the overall effectiveness characteristics of interactive content.

7. The method for optimizing scene-based interactive content based on virtual reality as described in claim 5, characterized in that, By integrating global and local interactive content effect features, the integrated interactive content effect evaluation result is obtained, including: Calculate the feature element similarity between the global interactive content effect features and the local interactive content effect features to obtain a feature element similarity set; The similarity set of the feature elements is normalized to construct an integrated interaction matrix; The integrated interaction matrix is ​​used to enhance the interaction of the local interactive content features, thereby obtaining the evaluation result of the integrated interactive content effect.

8. The method for optimizing scene-based interactive content based on virtual reality as described in claim 1, characterized in that, Based on the evaluation results of the integrated interactive content effect, the interactive content configuration of the virtual scene in the virtual-real structure network is optimized to obtain the virtual scene interactive content optimization results, including: Obtain the effect of the target interactive content, identify deviations from the evaluation results of the integrated interactive content effect, and obtain the direction and degree of deviation. The virtual scene is configured with directional interactive content optimized based on the deviation direction and degree to obtain the initial virtual scene interactive content optimization result; The data repository of the virtual-real structure network platform is invoked to predict the optimization effect of the initial virtual scene interaction content optimization result. Based on the prediction result, it is determined whether the target interaction content effect is met. If so, the initial virtual scene interaction content optimization result is taken as the virtual scene interaction content optimization result.

9. The method for optimizing scene-based interactive content based on virtual reality as described in claim 8, characterized in that, If not, the initial virtual scene interaction content optimization result is optimized again based on the target interaction content effect until the prediction result meets the target interaction content effect.

10. A virtual reality-based scenario-based interactive content optimization system, characterized in that: The system is used to implement the virtual reality-based scenario-based interactive content optimization method according to any one of claims 1 to 9, the system comprising: The platform construction module is used to build a virtual-real structure network platform for offline interactive scenarios, wherein the virtual-real structure network platform performs structured modeling of offline scenarios, virtual interactive content and user behavior; The interaction instance acquisition module is used to trigger the loading of the virtual reality interactive scene of the virtual and real structure network platform based on the user's scanning behavior, and to verify the user's real position through spatial perception. When the verification is successful, the interaction state in the virtual and real structure network platform is updated to obtain the activated virtual interaction instance. The evaluation result acquisition module is used to launch the virtual scene of the virtual-real structure network platform according to the activated virtual interaction instance, collect user behavior data during the virtual reality interaction process, and perform a two-dimensional evaluation of the interactive content effect based on local features and global features to obtain an integrated interactive content effect evaluation result. The optimization result acquisition module is used to optimize the interactive content configuration of the virtual scene in the virtual-real structure network based on the integrated interactive content effect evaluation result, and obtain the virtual scene interactive content optimization result.

Images