Augmented reality content recommendation and expressive presentation method based on user behavior perception
By constructing a state representation and adaptive weight mechanism for virtual characters, the position and orientation of virtual characters in the AR system are optimized, solving the problems of visual conflict and low information transmission efficiency in the AR system, and improving the user interaction experience and information transmission effect.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- BEIJING TECH & BUSINESS UNIV
- Filing Date
- 2025-12-12
- Publication Date
- 2026-06-26
Smart Images
Figure CN121685902B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the fields of computer vision and augmented reality, and in particular to an augmented reality content recommendation and expressive presentation method based on user behavior perception. Background Technology
[0002] With the rapid development of Augmented Reality (AR) technology in fields such as computer vision, graphics rendering, and human-computer interaction, it is gradually becoming more widespread in applications such as educational displays, digital museums, city navigation, and retail. AR systems enhance the spatial representation of information and the user's immersion by overlaying virtual content onto the real world, becoming an important means of information acquisition and interaction for the next generation. However, current mainstream AR systems still face limitations in their display strategies, restricting their universal adaptability in complex and open scenarios.
[0003] Current AR systems mostly employ a uniform template-driven visualization approach, emphasizing information density and visual neatness while lacking expressive modeling and stylistic diversity. Especially when multiple pieces of content coexist or complex scene transitions occur, occlusion, interference, or visual crowding can easily arise, affecting the user's efficiency in receiving important information. Furthermore, existing content layout strategies lack the ability to perceive user perspective stability, movement paths, and shifts in interest, failing to achieve dynamic content organization and visual guidance tailored to user behavior.
[0004] Therefore, how to dynamically optimize the layout of virtual characters, avoid visual conflicts, and improve the naturalness, harmony, and semantic expression of the layout, thereby enhancing the interactive experience and information transmission effect of AR systems, is a technical problem that urgently needs to be solved by those skilled in the art. Summary of the Invention
[0005] This invention provides an augmented reality content recommendation and expressive presentation method based on user behavior perception, which dynamically optimizes the layout of virtual characters, avoids visual conflicts, and improves the naturalness, coordination and semantic expressiveness of the layout, thereby enhancing the interactive experience and information transmission effect of the AR system.
[0006] On the one hand, the present invention provides an augmented reality content recommendation and expressive presentation method based on user behavior perception, which includes:
[0007] Construct a state representation for multiple virtual characters in an augmented reality scene; the state representation includes the three-dimensional spatial position, facing direction, and occlusion state of each virtual character;
[0008] A joint objective function is constructed to evaluate the rationality of the layout of virtual characters; the joint objective function includes a spatial cost function and an orientation cost function, wherein the spatial cost function is used to quantify the rationality of the spatial relationship between characters, and the orientation cost function is used to quantify the coordination and diversity of the characters' orientations;
[0009] Based on the attribute information of the virtual character, adaptive weights are generated through a pre-built weight prediction model, and the adaptive weights are used to adjust the cost terms related to the importance of the character in the spatial cost function.
[0010] Based on the joint objective function, the position and orientation of all virtual characters are iteratively optimized to obtain an optimized layout;
[0011] The optimized layout is presented expressively in augmented reality scenarios.
[0012] The present invention provides an augmented reality content recommendation and expressive presentation method based on user behavior perception. This method constructs a state representation of virtual characters, designs a joint optimization objective that integrates spatial rationality and orientation expressiveness, introduces an adaptive weight mechanism based on character attributes, and iteratively optimizes the position and orientation of all virtual characters based on the joint objective function to obtain an optimized layout. This optimized layout is then expressively presented in the augmented reality scene. This method achieves automatic optimization of character position and orientation in three-dimensional space, avoids visual conflicts, and improves the naturalness, coordination, and semantic expressiveness of the layout, thereby significantly enhancing the interactive experience and information transmission effect of the AR system. Attached Figure Description
[0013] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.
[0014] Figure 1 This is a flowchart illustrating the augmented reality content recommendation and expressive presentation method based on user behavior perception provided in an embodiment of the present invention.
[0015] Figure 2 This is a schematic diagram of the structure of the augmented reality content recommendation and expressive presentation system based on user behavior perception provided in an embodiment of the present invention. Detailed Implementation
[0016] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.
[0017] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this invention, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.
[0018] Figure 1 This is a flowchart illustrating the augmented reality content recommendation and expressive presentation method based on user behavior perception provided in this embodiment of the invention.
[0019] like Figure 1 As shown, the augmented reality content recommendation and expressive presentation method based on user behavior perception provided in this embodiment of the invention mainly includes the following steps:
[0020] 101. Construct a state representation of multiple virtual characters in an augmented reality scene; the state representation includes the three-dimensional spatial position, facing direction, and occlusion state of each virtual character;
[0021] 102. Construct a joint objective function for evaluating the rationality of virtual character layout; the joint objective function includes a spatial cost function and an orientation cost function, wherein the spatial cost function is used to quantify the rationality of spatial relationships between characters, and the orientation cost function is used to quantify the coordination and diversity of character orientations;
[0022] 103. Based on the attribute information of the virtual character, generate adaptive weights through a pre-built weight prediction model, and use the adaptive weights to adjust the cost terms related to the importance of the character in the spatial cost function;
[0023] 104. Based on the joint objective function, the positions and orientations of all virtual characters are iteratively optimized to obtain an optimized layout;
[0024] 105. Expressive presentation of the optimized layout in augmented reality scenarios.
[0025] Virtual characters refer to digital images presented in augmented reality scenarios, which may have specific attributes, functions, and interaction logic, such as virtual guides, virtual audiences, and virtual exhibit icons.
[0026] The state representation is a set of structured descriptions of the core features of the virtual character in the scene, which is used for subsequent layout optimization and interaction reasoning.
[0027] The three-dimensional spatial position refers to the specific coordinates of the virtual character in the three-dimensional coordinate system of the augmented reality scene, which is used to locate the character's spatial position.
[0028] The facing direction refers to the direction the virtual character is pointing, which affects the user's visibility and interactive perception of the character.
[0029] Occlusion status indicates whether a virtual character is occluded by other characters, scene objects, or boundaries, and is in an invisible or partially visible state.
[0030] The joint objective function is a mathematical model that comprehensively evaluates the advantages and disadvantages of virtual character layout. By integrating spatial and orientation-related cost terms, it provides a quantitative basis for layout optimization.
[0031] The spatial cost function is used to measure the rationality of spatial relationships between virtual characters and between characters and the scene, quantifying problems such as collisions, boundary crossings, and uneven distribution.
[0032] The orientation cost function is used to measure the coordination and diversity of virtual character orientations, ensuring that the character orientation conforms to the interaction logic and visual expression requirements.
[0033] Attribute information consists of inherent characteristic data of virtual characters, including categorical characteristics (such as occupation and character type) and numerical characteristics (such as appearance frequency and seniority level).
[0034] Adaptive weights are importance coefficients dynamically generated based on character attributes. They can be adjusted according to the scene and character status to optimize layout priority.
[0035] Specifically, firstly, for each target virtual character in the current AR scene, its three-dimensional spatial position is determined, i.e., the specific spatial coordinates of each character are located using the scene coordinate system; secondly, the facing direction of each character is determined, using a preset reference direction in the scene (such as the user's direct view or the direction of the scene center) as a benchmark; and thirdly, it is determined whether each character is occluded, using methods such as scene object contour detection and character spatial overlap judgment to mark the occlusion state (visible or invisible) of the character, ultimately forming a complete state representation of each virtual character including its three-dimensional spatial position, facing direction, and occlusion state. The state representation of each virtual character can be denoted as... The overall set can be represented by equation (1):
[0036] (1)
[0037] in, Absolute position coordinates representing a three-dimensional spatial location, usually in meters (m), used to determine its specific geographical location in the scene; The angle between the direction the speaker is facing and the direction directly in front of them is expressed in radians (rad). Its value affects the visibility and interactivity of the content for the user. This is a binary flag indicating whether the character is currently occluded; 1 indicates visibility, and 0 indicates occlusion or invisibility.
[0038] Next, a joint objective function is constructed to evaluate the rationality of the layout. Among them, the spatial cost function mainly quantifies the rationality of the spatial relationships between characters and between characters and the scene, such as determining whether characters collide, whether they are close to the scene boundary, whether they are reasonably distributed around the scene center, and whether they uniformly cover the scene area; the orientation cost function focuses on the coordination and diversity of character orientations, ensuring that the character orientations meet the scene interaction requirements (such as facing the user or the interaction focus), while avoiding visual monotony caused by all characters having completely identical orientations.
[0039] Next, attribute information for each virtual character is collected, such as the character's profession, functional role in the scene, historical appearance frequency, and seniority level. After preprocessing this attribute information, it is input into a pre-trained weight prediction model. The model automatically outputs the importance weights for each character based on the attribute features, i.e., adaptive weights. These weights are then applied to the spatial cost function to adjust the cost terms related to the character's importance. For example, characters with higher importance are more likely to be placed closer to the scene center, reducing their probability of being occluded.
[0040] Then, based on the joint objective function constructed above, iterative optimization of the positions and orientations of all virtual characters is initiated. In each iteration, according to the calculation results of the joint objective function, the 3D spatial positions of some or all virtual characters are adjusted (e.g., moving overlapping characters, adjusting important characters that are off-center, and dispersing overly dense characters) and their facing directions (e.g., correcting the orientation of characters that are away from the interaction focus, and increasing the diversity of orientations). The cost of the joint objective function is then recalculated to determine if the optimization objective has been achieved. If not, iterative adjustments continue until the cost stabilizes within a preset range or the required number of iterations is met, resulting in the final optimized layout.
[0041] Finally, the optimized virtual character layout is presented expressively within an augmented reality scene. During this expressive presentation, visual aesthetic principles can be followed, allowing for fine-tuning of attributes such as character size and color to ensure the layout not only meets the optimization goals but also enhances the user's visual experience and information reception efficiency.
[0042] This embodiment of the augmented reality content recommendation and expressive presentation method based on user behavior perception constructs a state representation of virtual characters, designs a joint optimization objective that integrates spatial rationality and orientation expressiveness, introduces an adaptive weight mechanism based on character attributes, and iteratively optimizes the position and orientation of all virtual characters based on the joint objective function to obtain an optimized layout. This optimized layout is then expressively presented in the augmented reality scene, achieving automatic optimization of character position and orientation in three-dimensional space, avoiding visual conflicts, and improving the naturalness, coordination, and semantic expressiveness of the layout, thereby significantly enhancing the interactive experience and information transmission effect of the AR system.
[0043] In some embodiments, the process of constructing a joint objective function for evaluating the rationality of virtual character layout includes:
[0044] Construct the spatial cost function, which is the sum of the collision cost function, the boundary cost function, the center cost function, the distribution cost function, and the region cost function;
[0045] The orientation cost function is constructed as the sum of the deviation cost function and the diversity cost function. The deviation cost function calculates the deviation between each virtual character's current orientation and its ideal orientation, which is determined by the virtual character's position relative to a preset point of interest. The diversity cost function encourages diversity in the distribution of orientations for all virtual characters.
[0046] The collision cost function, calculated based on the intersection-union ratio (IUR) of the bounding boxes of the virtual characters, is a sub-function used to quantify the degree of spatial overlap between virtual characters. The boundary cost function penalizes virtual characters close to the scene boundary, preventing characters from exceeding the user's visual range or being at the scene edge, thus affecting the experience. The center cost function encourages virtual characters to move closer to the scene center according to their corresponding importance weights, guiding them to increase the visual priority of important characters. The distribution cost function encourages virtual characters to be evenly distributed in space, preventing excessive clustering or dispersion of characters. The region cost function encourages virtual characters to cover multiple segmented regions of the scene, ensuring full utilization of the scene space.
[0047] A character bounding box is a virtual bounding box constructed for a virtual character, used to simplify collision detection calculations for the character's spatial position. The intersection-union ratio (IU) is the ratio of the volume of the overlapping portion of two character bounding boxes to the total volume of the two bounding boxes, used to determine whether characters collide and the degree of collision.
[0048] The deviation cost function quantifies the difference between the virtual character's current orientation and its ideal orientation, calculating the deviation of each virtual character's current orientation from its ideal orientation. The diversity cost function encourages a diverse distribution of virtual character orientations, avoiding visual monotony caused by a single orientation. The ideal orientation is the optimal orientation determined based on the virtual character's position relative to a preset point of interest, ensuring that the character's orientation conforms to the interaction logic.
[0049] The preset focus points are the core interaction locations that are pre-defined in the scene, such as the user's current location, the central point of the scene's functions, and the location of important virtual objects.
[0050] Specifically, the spatial cost function comprehensively quantifies the rationality of spatial layout through the superposition of multi-dimensional sub-functions. See formula (2) for the formula:
[0051] (2)
[0052] in, Represents the collision cost function; The boundary cost function; The cost function is centered. Represents the distributed cost function; This is the region cost function.
[0053] The collision cost function penalizes the case where any two virtual characters overlap spatially, using the intersection-over-union ratio (IoU) between the character bounding boxes as the criterion. The specific form is shown in equation (3):
[0054] (3)
[0055] in, Represents virtual characters Character wrap-around box; Represents virtual characters Role-enclosing box; I( ) represents an indicator function, which is 1 if the condition is true and 0 otherwise. N represents the number of virtual characters.
[0056] The cost penalty for the boundary cost function is applied to characters closer to the scene boundary, as shown in equations (4) and (5):
[0057] (4)
[0058] (5)
[0059] in, and This represents the half-width of the scene boundary in the x and z directions; Represents virtual characters The coordinates in the x-direction; Represents virtual characters The coordinates in the z-direction; The edge buffer distance is usually set to 2m (adjusted according to the specific space size).
[0060] The cost of the center cost function encourages important characters to move closer to the center of the scene, as shown in equation (6):
[0061] (6)
[0062] in, For the role Importance weights; The coordinates of the scene center; This indicates Euclidean distance.
[0063] The cost of the distributed cost function encourages uniform distribution among roles and suppresses role clustering, as shown in equation (7):
[0064] (7)
[0065] This involves calculating the standard deviation of the distance between all character pairs; a smaller standard deviation indicates greater uniformity. Let represent the Euclidean distance between the i-th virtual character and the j-th virtual character. This means iterating through all virtual character pairs where the index i is less than the index j. The purpose is to ensure that each unordered virtual character pair is calculated only once (to avoid repeatedly calculating the distance of the same virtual character pair).
[0066] The cost of the region cost function divides the scene into M fixed grid regions. If a region is not occupied by any character, a penalty is incurred, as shown in formula (8):
[0067] (8)
[0068] in, This represents the number of characters contained in the k-th region; This is an indicator function.
[0069] The orientation cost function achieves coordination and enrichment of orientation by combining deviation penalty and diversity incentive, as shown in equation (9):
[0070] (9)
[0071] in, The bias cost function; For diversity cost function.
[0072] The ideal orientation of each character It is determined by its vector relative to the preset focus point (such as the center of the stage), as shown in equation (10):
[0073] (10)
[0074] The orientation deviation is calculated by taking the inverse cosine of the cosine angle and then calculating its absolute value. The smaller the value, the closer the character's orientation is to the stage direction, as shown in formula (11):
[0075] (11)
[0076] The diversity cost function encourages a more diverse distribution of orientations for all characters, preventing monotonous performance caused by uniform orientations. It can be defined as the negative of the orientation standard deviation, as shown in formula (12).
[0077] (12)
[0078] Here, the standard deviation of all character orientations is calculated using a diversity cost function. A larger standard deviation indicates a more diverse orientation distribution, a smaller function value (negative value), and a lower cost; conversely, a smaller standard deviation indicates a more concentrated orientation, a larger function value, and a higher cost. In some embodiments, entropy or variance can be used instead of standard deviation as a diversity measure, while maintaining the same core logic, and will not be illustrated further here.
[0079] In some embodiments, the process of generating adaptive weights using a pre-built weight prediction model based on the attribute information of a virtual character may include: embedding and encoding categorical features in the attribute information to obtain a categorical feature vector; normalizing numerical features to obtain a numerical feature vector; concatenating the categorical feature vector and the numerical feature vector to obtain a unified feature vector; and inputting the unified feature vector into the weight prediction model for weight prediction to obtain the adaptive weights.
[0080] Categorical features are classification features in virtual character attribute information that cannot be directly quantified, such as the character's occupation type, function category, and status level. Embedding encoding is a processing method that converts categorical features into low-dimensional dense vectors that can be processed by computers, preserving the semantic relationships of the features. Numerical features are features in virtual character attribute information that can be directly quantified, such as the number of times they appear on screen, their dwell time, and their experience score. The categorical feature vector is a low-dimensional vector obtained through embedding encoding, used to represent categorical features. The numerical feature vector is a vector obtained through normalization, used to represent numerical features. The unified feature vector is a single vector formed by combining the categorical feature vector and the numerical feature vector according to a preset rule, used as the input to the weight prediction model.
[0081] Specifically, the attribute information of the virtual characters is first classified into categorical features and numerical features. For categorical features, such as the character's occupation (e.g., narrator, audience member, host), function (e.g., interactive, demonstration), and status level (e.g., advanced, intermediate, beginner), embedding encoding is used. Through a pre-trained embedding layer model, each categorical feature is mapped to a low-dimensional, dense categorical feature vector. This vector preserves the semantic relationships between different categories (e.g., the similarity between the category vectors of a host and a narrator is higher than that between a host and an ordinary audience member), facilitating subsequent feature learning by the model. For numerical features, such as the character's historical appearance count, dwell time in the scene, number of user interactions, and seniority score, normalization is performed. Based on the value range of the numerical features, a min-max normalization method is used to map all numerical feature values to a fixed range of 0-1, eliminating model training bias caused by differences in units of measurement (e.g., dozens of appearances versus several minutes of dwell time), resulting in a standardized numerical feature vector.
[0082] After processing the categorical and numerical features separately, the resulting categorical and numerical feature vectors are concatenated in a preset order. For example, all elements of the categorical feature vector are arranged first, followed by all elements of the numerical feature vector, forming a unified feature vector containing all attribute feature information, ensuring that this vector can comprehensively reflect the attribute features of the virtual character.
[0083] The constructed unified feature vector is input into a pre-trained weight prediction model. This model employs a lightweight deep neural network structure and has been trained using a large amount of virtual character attribute data labeled with importance weights. The model extracts and maps features from the unified feature vector and outputs the corresponding weight values. These weight values are adaptive weights used to adjust the spatial cost function, accurately reflecting the importance level of the virtual character in the current scene layout.
[0084] Furthermore, the process of obtaining adaptive weights includes: inputting the unified feature vector into the weight prediction model to predict the weights and obtain the predicted weights for the current optimized layout; and using a smooth update mechanism to dynamically calibrate the predicted weights to obtain the adaptive weights.
[0085] Specifically, the first step is to obtain the predicted weights for the current optimized layout. The unified feature vector of each virtual character is input into the weight prediction model one by one. Based on the features of the current scene and the character attributes, the model outputs the initial weight value corresponding to each character, which is the predicted weight of the current optimized layout. This weight reflects the model's preliminary judgment on the importance of the current character.
[0086] Because AR scenes are dynamic, the state of virtual characters may change due to user behavior, scene transitions, and other factors. Relying solely on the predicted weights for the current iteration may lead to fluctuations or deviations. Therefore, a smoothing update mechanism is needed to dynamically calibrate the predicted weights. The core of the smoothing update mechanism is to adjust the predicted weights for the current iteration by combining historical weight information or preset stability rules, avoiding significant fluctuations in weights across different optimization iterations. For example, if the predicted weights for the current iteration differ significantly from those in previous optimization iterations, the smoothing mechanism will appropriately reduce the adjustment magnitude of the predicted weights for the current iteration, ensuring that the calibrated weights can respond to changes in the current scene while maintaining a certain level of stability.
[0087] Through the above smooth update and dynamic calibration process, the initial prediction weights are adjusted to weight values that are more in line with the actual needs of the scenario and have stronger stability. These weight values are the adaptive weights used to adjust the spatial cost function.
[0088] In a specific implementation, the training process of the weight prediction model includes:
[0089] a. Construct a labeled dataset for weight prediction training, derived from typical scenarios such as conference presentations, interviews, and virtual theaters. Define a set of historical roles. Each historical character It has a set of attribute vectors:
[0090] Each attribute This represents a structured feature of a historical role, such as occupation type, organizational level, historical participation frequency, role seniority, and visual exposure; attributes include categorical (e.g., job title) and numerical (e.g., number of appearances) information. Domain experts label each role in the dataset with "importance," forming the true weights. , used as a training supervision signal.
[0091] b. The attribute information of historical characters is characterized by feature encoding and normalization to obtain the corresponding unified feature vector. For details, please refer to the above record, which will not be repeated here.
[0092] c. Weighted prediction model training: A lightweight deep neural network (such as ResNet-18) is used as the basic prediction structure, and the L1 norm loss is used as the supervision signal, as shown in equation (13):
[0093] (13)
[0094] in, This represents the domain expert's label for the importance of the role (value range [0,1]). These are the predicted weights for the model. The model is trained using a large amount of labeled data to minimize the absolute deviation between the predicted and actual weights, thus enabling the model to have accurate weight prediction capabilities.
[0095] In some embodiments, the process of dynamically calibrating the predicted weights using a smooth update mechanism to obtain the adaptive weights may include:
[0096] Maintain a weight cache pool, which stores the historical weights corresponding to each virtual character in the historical optimized layout; for each virtual character in the current optimized layout, obtain the historical weight of each virtual character from the weight cache pool, and dynamically fuse the predicted weight with the historical weight to obtain the fused weight; use the fused weight as the adaptive weight, and update the adaptive weight to the weight cache pool for smooth updates in subsequent rounds.
[0097] The weight cache pool is a storage unit used to store the adaptive weights of virtual characters in historical optimization rounds, and supports reading, writing and updating operations of weights.
[0098] In this embodiment, when implementing the smooth update mechanism, a dedicated weight cache pool is first maintained. This cache pool is categorized and stored according to the optimization round and the virtual character identifier, saving the historical weights (i.e., the adaptive weights after dynamic calibration in that round) corresponding to each virtual character in each round of historical optimization layout, ensuring that the past weight information of each character can be quickly retrieved.
[0099] For each virtual character in the current optimized layout, the historical weight of that character in all historical optimization rounds is retrieved from the weight cache pool based on the character's identifier. If the character is newly added to the scene and has no historical weight record, a base weight value is assigned by default as a historical weight for calculation.
[0100] Subsequently, the predicted weights obtained in the current iteration are dynamically fused with the extracted historical weights. The fusion process must balance the adaptability of the predicted weights to the current scenario with the stability of the historical weights. For example, based on the time decay characteristics of weights, recent historical weights are given a higher reference weight, while older historical weights are given a lower reference weight, while retaining the core influence of the predicted weights. The fused weights are obtained through weighted calculation. The calculated fused weights are directly used as the adaptive weights for the current round, used to adjust the spatial cost function. Simultaneously, these adaptive weights are updated in the weight cache pool according to the role identifier and the current optimization round, overwriting or supplementing the historical weight records for that role, providing the latest historical data support for smooth updates in subsequent optimization rounds.
[0101] Furthermore, the process of obtaining the fusion weights may include:
[0102] The absolute difference between the predicted weight and the mean of all historical weights is determined; based on the absolute difference and a preset sensitivity threshold, an influence factor is assigned to the predicted weight and each historical weight; the product of the predicted weight and each historical weight with their respective influence factors is summed to obtain the fused weight; wherein, the influence factor includes a first influence factor for the predicted weight and a second influence factor assigned to each historical weight, the first sum of the first influence factor and all second influence factors being 1; and the second influence factor of each historical weight decreases as the storage time increases; if the absolute difference is greater than or equal to the preset sensitivity threshold, the first influence factor is greater than the second sum of all second influence factors; if the absolute difference is less than the preset sensitivity threshold, the first influence factor is less than the second sum of all second influence factors.
[0103] The preset sensitivity threshold is a pre-defined critical value used to judge the degree of deviation of the predicted weights, determined based on scenario requirements and optimization experience. The influence factor is a coefficient used to adjust the proportion of predicted weights and historical weights in dynamic fusion, including a first influence factor and a second influence factor. The first influence factor is the influence factor assigned to the predicted weights, determining the contribution of the predicted weights in the fusion. The second influence factor is the influence factor assigned to each historical weight, determining the contribution of the corresponding historical weight in the fusion.
[0104] Specifically, the first step is to calculate the average of all historical weights, which is to perform an arithmetic average of all historical weights of the virtual character in the weight cache pool to obtain a value that reflects the overall level of historical weights.
[0105] Next, the absolute difference between the current predicted weight and the average of those weights is calculated. This difference is used to determine the degree of deviation between the current predicted weight and the historical weights. At the same time, a preset sensitivity threshold is retrieved from the system. This threshold is pre-set based on factors such as the dynamic characteristics of the AR scene and the need for optimized stability. It is used to distinguish whether the deviation of the predicted weight is a normal fluctuation or a significant change.
[0106] Based on the comparison between the absolute difference and a preset sensitivity threshold, a first influence factor is assigned to the predicted weights, and a second influence factor is assigned to each historical weight. The sum of the first influence factor and all second influence factors is fixed at 1 to ensure that the calculation of the fused weights conforms to the probability distribution rule. Simultaneously, the second influence factor for each historical weight follows a time decay rule; the longer a historical weight has been stored, the smaller its corresponding second influence factor, meaning that more recent historical weights have a greater impact on the fusion result.
[0107] If the absolute difference is greater than or equal to the preset sensitivity threshold, it indicates that there is a significant difference between the current predicted weight and the historical weight, which may be due to a significant change in the scenario (such as a change in role function or a shift in user focus). In this case, a higher first influence factor is assigned, and the first influence factor is greater than the sum of all second influence factors, so that the predicted weight plays a dominant role in the fusion process to quickly respond to changes in the scenario. If the absolute difference is less than the preset sensitivity threshold, it indicates that the predicted weight deviates little from the historical weight, and the scenario is in a relatively stable state. In this case, a lower first influence factor is assigned, and the first influence factor is less than the sum of all second influence factors, so that the historical weight plays a major role in the fusion process to ensure the stability of the weight.
[0108] Finally, the predicted weights are multiplied by the first influence factor, each historical weight is multiplied by its respective second influence factor, and all the product results are summed to obtain the fusion weight, which is used as the adaptive weight for the current round.
[0109] In some embodiments, the iterative optimization process may include:
[0110] By using a neural network model based on an attention mechanism (such as the Transformer model), an initial layout of the global structure is generated based on the semantic features of the current augmented reality scene and the initial state of the virtual character, which serves as the current layout state.
[0111] Based on the current layout, all virtual characters are divided into multiple spatial function groups according to spatial proximity or functional similarity.
[0112] In each iteration, a group-level action is sampled from the preset action space and executed for each of the spatial function groups. The group-level action is used to coordinate the position and orientation of all virtual characters in the group in three-dimensional space.
[0113] Based on the updated layout state and the joint objective function, a group-level reward is calculated for each spatial function group that has performed a group-level action; the group-level reward is the negative of the sum of the spatial cost function and the orientation cost function corresponding to the spatial function group.
[0114] An optimization algorithm based on policy gradient is adopted to update the action selection strategy of each spatial function group according to the group-level reward, and the updated layout state is used as the current layout state for the next iteration until the preset convergence condition is met, and the final optimized layout is output.
[0115] The initial state is the original state representation of the virtual character before optimization begins, including the initial three-dimensional spatial position, facing direction, and occlusion state.
[0116] The initial layout is an initial character layout with global structural rationality generated by a neural network model, providing a foundation for subsequent iterative optimization.
[0117] Spatial function groups are sets of roles divided according to the spatial proximity or functional similarity of virtual characters, which facilitates coordinated adjustments.
[0118] Spatial proximity refers to the degree of closeness between virtual characters in three-dimensional space; the closer the distance, the higher the spatial proximity.
[0119] Functional similarity refers to the degree of similarity in the functions performed by virtual characters in a scene. The closer the functions are, the higher the functional similarity.
[0120] The preset motion space is a predefined set of actions used to adjust the position and orientation of a virtual character, including actions such as translation and rotation.
[0121] Group-level actions are sampled from a preset action space and are used to coordinate and adjust the roles within the entire functional group.
[0122] Group-level rewards are indicators used to evaluate the layout optimization effect after a spatial functional group performs group-level actions, and are associated with the cost of the joint objective function.
[0123] In detail, the initial layout is generated. An attention-based neural network model is used, which can automatically focus on the key semantic features of the current AR scene (e.g., when the scene type is a virtual meeting, it focuses on features such as the central area of the meeting and the distribution of seats; when the scene type is a city tour, it focuses on features such as the location of attractions and the tour route). At the same time, it combines the initial state of all virtual characters (initial position, orientation, occlusion state), and through the model's feature extraction and layout generation capabilities, outputs an initial layout with a globally reasonable structure. This layout avoids obvious collision and boundary crossing problems in the initial state and serves as the current layout state for subsequent iterative optimization.
[0124] The initial layout aims to provide a well-structured initial solution for subsequent optimization processes, thereby reducing invalid search paths and shortening convergence time. Scene semantic features may include, for example, stage shape, obstacle locations, speaker locations, etc. The initial coarse layout scheme can be spatial regularity initialization or uniform distribution. After structured encoding, it is input into the Transformer model, and the initial estimated state is output, as shown in equation (14):
[0125] (14)
[0126] in, This represents the initial optimization point, i.e., the initial state of the virtual character.
[0127] Based on the generated current layout state, all virtual characters are divided into spatial functional groups. The grouping method can be either spatial proximity or functional similarity: if spatial proximity is selected, characters that are close together in 3D space are grouped together; if functional similarity is selected, characters that perform the same or similar functions in the scene (e.g., both are narrators, both are audience members) are grouped together. By grouping these characters, the large-scale character layout optimization problem is decomposed into multiple smaller-scale group optimization problems, reducing the optimization complexity.
[0128] Specifically, to reduce the joint search dimensions, the N virtual characters are divided into G spatial function groups. The criteria include: spatial proximity (such as location-based clustering); and functional similarity (such as identity, role attributes, etc.).
[0129] The formula for defining the local state of the g-th spatial functional group is as shown in equation (15):
[0130] (15)
[0131] The formula for the overall global state is as shown in equation (16):
[0132] (16)
[0133] in, This represents the "union" operation of sets, which merges all local state sets from group 1 to group G (since each role belongs to only one group, there are no duplicates after merging).
[0134] In the iterative optimization process, in each iteration, a group-level action is sampled from the preset action space for each spatial function group. The preset action space contains various collaborative actions for adjusting character positions and orientations, such as overall translation of characters within the group, overall rotation of characters, and adjustment of spacing between characters within the group. After executing the group-level action, the 3D spatial positions and orientations of all virtual characters within the group will be collaboratively adjusted according to the action rules, generating an updated layout state.
[0135] Specifically, for each group of spatial functions Define the action as a fine-tuning operation in three-dimensional space. The corresponding formula is shown in equation (17):
[0136] (17)
[0137] in, This represents the fine-tuning of the spatial function group's position in the x-axis direction in three-dimensional space (i.e., the change in the x-coordinate of the character within the group). This represents the positional adjustment of a spatial function group in the z-axis direction in three-dimensional space (i.e., the change in the z-coordinate of a character within the group). This represents the amount of fine-tuning of the orientation of all characters (i.e., the change in the character's orientation angle).
[0138] Each group of fine-tuning operations Determined by the policy function, its corresponding formula is shown in equation (18):
[0139] (18)
[0140] in, πg ( ) represents the strategy function corresponding to the g-th spatial function group, which calculates and outputs the action to be performed by the group based on the current state of the group. This represents the local state of the g-th spatial functional group.
[0141] Based on the updated layout state, and combined with the aforementioned joint objective function, the spatial cost function value and orientation cost function value corresponding to each spatial function group are calculated. The sum of the two cost function values and the inverse of the sum are taken to obtain the group-level reward for that spatial function group. The higher the group-level reward value, the better the layout optimization effect after the group performs the current group-level action.
[0142] Specifically, to measure the effect of each set of actions, the optimized cost function of that set is calculated and its negative value is taken as the reward. The corresponding calculation formula is shown in equation (19):
[0143] (19)
[0144] The global reward is the sum of the rewards for all groups, as shown in formula (20):
[0145] (20)
[0146] An optimization algorithm based on policy gradients is employed to adjust the action selection strategy of each spatial function group according to its group-level reward. For example, if a group receives a high group-level reward after performing a certain group-level action, the optimization algorithm will increase the probability of that group selecting that type of action in subsequent iterations; conversely, it will decrease the selection probability. Simultaneously, the updated layout state is used as the current layout state for the next iteration, and the process of group action sampling, execution, reward calculation, and policy updating is repeated.
[0147] Specifically, group-level strategy optimization algorithms (such as GRPO) can be used to update model parameters. The gradient update formula is as shown in equation (21):
[0148]
[0149] The strategy parameters are updated as follows:
[0150]
[0151] The iteration process terminates when the preset convergence condition is met. The convergence condition can be set as follows: the cost of the joint objective function remains stable for multiple consecutive iterations (i.e., the decrease in the joint objective function is less than a threshold), or the number of iterations reaches a preset upper limit. The output layout state at this time is the final optimized layout.
[0152] In practical applications, if virtual characters that closely match the user's real-time behavior and areas of interest are not effectively selected, the relevance of the recommended content may be insufficient, affecting the immersive experience of the user and the final effect of layout optimization.
[0153] To address this, the present invention provides the following technical solution:
[0154] Before constructing state representations for multiple virtual characters in an augmented reality scene, the following steps are also included:
[0155] Real-time acquisition of user behavior perception data in augmented reality scenarios; the behavior perception data includes the user's gaze direction, head posture, interaction actions, and dwell time;
[0156] Based on the behavioral perception data, the user's real-time attention area in the current scene is estimated through a pre-trained visual attention mechanism or a deep neural network model.
[0157] Calculate the correlation score between each candidate virtual character and the real-time attention area;
[0158] Candidate virtual characters are sorted according to the correlation score, and multiple virtual characters with scores higher than a preset threshold are selected.
[0159] Specifically, the AR device first collects user behavior data in real time through sensors (such as cameras, gyroscopes, gesture recognition modules, etc.), including the user's gaze direction (obtained through eye-tracking technology), head posture (detected by the gyroscope to measure the pitch and yaw angles of the head), interactive actions (such as the user clicking on a virtual character with gestures, asking for relevant information by voice, etc.), and the duration of the user's stay in different areas of the scene.
[0160] The collected behavioral perception data is then input into a pre-trained visual attention mechanism or deep neural network model. This model has been trained using a large number of user behavior samples and can determine the user's attention intent based on the characteristics of the behavioral data. For example, if a user's gaze is continuously directed towards a certain area of the scene, their head posture is facing that area, and they linger there for a relatively long time, the model will identify that area as the user's real-time attention area. If the user clicks on a virtual character through an interactive action, the model will identify the area where the character is located and related areas as the real-time attention area.
[0161] Then, for all candidate virtual characters in the AR scene, the relevance score between each character and the real-time attention area is calculated. The calculation of the relevance score needs to comprehensively consider factors such as the spatial distance between the character and the attention area, the matching degree between the character's semantics and the user's interests, and the consistency between the character's orientation and the user's line of sight. The specific calculation method will be explained in detail in the following content.
[0162] Finally, based on the calculated relevance scores, all candidate virtual characters are sorted in descending order. The higher the score, the stronger the correlation with the user's real-time focus area. A preset threshold is set, determined based on factors such as the total number of virtual characters in the scene and user attention needs. Virtual characters with scores higher than this preset threshold after sorting are selected. These characters are the target virtual characters for subsequent state representation construction, ensuring that subsequent layout optimization focuses on the core characters that users are concerned with, thus improving the targeting of the layout.
[0163] In a specific implementation, the process of calculating the relevance score includes:
[0164] The system obtains the spatial distance between the candidate virtual character's location and the real-time attention area, the matching degree between the candidate virtual character's semantic tags and the user's historical interests, and the angle between the candidate virtual character's orientation and the user's line of sight. Specifically, the Euclidean distance between the candidate virtual character's location and the geometric center of the real-time attention area is calculated as the spatial distance; the semantic tags of the candidate virtual character are compared with interest tags extracted from the user's historical behavior data, and their overlap is calculated as the matching degree; the minimum angle between the candidate virtual character's facing direction vector and the user's line of sight direction vector is calculated as the angle.
[0165] A spatial attenuation factor is generated based on the spatial distance, an orientation correction coefficient is generated based on the included angle, and a semantic base score is generated based on the matching degree. Specifically, according to a preset attenuation function, the spatial distance is mapped to a value between zero and one as the spatial attenuation factor; the closer the distance, the higher the mapped spatial attenuation factor value. According to a preset mapping rule, the included angle is mapped to a value between zero and one as the orientation correction coefficient; the smaller the included angle, the more consistent the character's orientation is with the user's line of sight, and the higher the mapped orientation correction coefficient value. Based on a preset scoring rule, an initial semantic base score is assigned to the candidate virtual character according to the degree of matching between the semantic tag and the user's historical interests; the higher the degree of matching, the higher the assigned semantic base score.
[0166] The spatial attenuation factor and the orientation correction coefficient are combined to obtain the real-time attention gain coefficient; specifically, the spatial attenuation factor and the orientation correction coefficient are multiplied to obtain the real-time attention gain coefficient.
[0167] The semantic base score is modulated using the real-time attention gain coefficient to obtain the modulated semantic interest score; specifically, the semantic base score is multiplied by the real-time attention gain coefficient to obtain the modulated semantic interest score.
[0168] The modulated semantic interest score is weighted and combined with the basic visibility score calculated based on spatial distance to output the final relevance score; where the closer the spatial distance, the higher the basic visibility score.
[0169] Specifically, the first step is to obtain three core evaluation metrics: spatial distance, matching degree, and angle. For spatial distance, the geometric center coordinates of the real-time focus area are first determined, and the Euclidean distance (i.e., straight-line distance) between the three-dimensional spatial position of each candidate virtual character and the geometric center is calculated. This distance is used for subsequent calculations.
[0170] For matching degree, semantic tags (such as the type of exhibit and functional attributes corresponding to the character) are extracted for each candidate virtual character. At the same time, interest tags are extracted from the user's historical behavior data (such as exhibits that have been followed and characters that have been interacted with). The semantic tags and interest tags are compared, and the proportion of the number of identical tags to the total number of tags is counted. This proportion is the matching degree.
[0171] For the included angle, the facing direction of the candidate virtual character and the user's line of sight are converted into direction vectors respectively, and the minimum angle between the two direction vectors is calculated. This angle is the included angle used for subsequent calculations.
[0172] Next, corresponding coefficients and scores are generated based on the three indicators mentioned above. For spatial distance, it is mapped to a spatial attenuation factor between 0 and 1 using a preset attenuation function; the closer the distance, the higher the spatial attenuation factor, and vice versa. For angle, it is mapped to an orientation correction coefficient between 0 and 1 using a preset mapping rule; the smaller the angle, the more consistent the character's orientation is with the user's line of sight, and the higher the orientation correction coefficient, and vice versa. For matching degree, a semantic base score is assigned according to a preset scoring rule; the higher the matching degree, the higher the semantic base score. For example, when the matching degree is 100%, the semantic base score is full; when the matching degree is 0, the semantic base score is 0.
[0173] The generated spatial attenuation factor is then fused with the orientation correction coefficient by multiplication to obtain the real-time attention gain coefficient, which comprehensively reflects the spatial correlation and orientation consistency between the character and the user's real-time attention area.
[0174] Then, the semantic base score is modulated using the real-time attention gain coefficient, and the modulated semantic interest score is obtained by multiplication. This score reflects both the relationship between the role and the user's historical interests and the user's current real-time attention status.
[0175] Finally, a basic visibility score is calculated based on spatial distance. The closer the spatial distance, the better the basic visibility of the character, and the higher the basic visibility score. The modulated semantic interest score and the basic visibility score are then weighted and combined according to preset weights. For example, each is assigned a preset weight coefficient, multiplied, and summed to obtain the final relevance score.
[0176] Based on the same general inventive concept, this invention also protects an augmented reality content recommendation and expressive presentation system based on user behavior perception. The augmented reality content recommendation and expressive presentation system based on user behavior perception provided by this invention will be described below. The augmented reality content recommendation and expressive presentation system based on user behavior perception described below can be referred to in correspondence with the augmented reality content recommendation and expressive presentation method based on user behavior perception described above.
[0177] Figure 2 This is a schematic diagram of the structure of the augmented reality content recommendation and expressive presentation system based on user behavior perception provided in an embodiment of the present invention, as shown below. Figure 2 As shown, the augmented reality content recommendation and expressive presentation system based on user behavior perception in this embodiment includes a first construction module 21, a second construction module 22, a generation module 23, an optimization module 24, and a display module 25.
[0178] In a specific implementation, the first construction module 21 is used to construct the state representation of multiple virtual characters in the augmented reality scene; the state representation includes the three-dimensional spatial position, facing direction and occlusion state of each virtual character;
[0179] The second construction module 22 is used to construct a joint objective function for evaluating the rationality of the layout of virtual characters; the joint objective function includes a spatial cost function and an orientation cost function, the spatial cost function is used to quantify the rationality of the spatial relationship between characters, and the orientation cost function is used to quantify the coordination and diversity of the character orientations;
[0180] The generation module 23 is used to generate adaptive weights based on the attribute information of the virtual character through a pre-built weight prediction model, and to use the adaptive weights to adjust the cost terms related to the importance of the character in the spatial cost function;
[0181] Optimization module 24 is used to iteratively optimize the position and orientation of all virtual characters based on the joint objective function to obtain an optimized layout;
[0182] Display module 25 is used to expressively present the optimized layout in an augmented reality scene.
[0183] It should be noted that all relevant information that may be involved in the various embodiments of the present invention is processed in strict accordance with the requirements of laws and regulations, following the principles of legality, legitimacy, and necessity, based on the reasonable purpose of the business scenario, and is information that users actively provide or generate during the use of the product / service, as well as information obtained with user authorization.
[0184] The information processed by this invention may vary depending on the specific product / service scenario and should be based on the specific scenario in which the user uses the product / service. This may involve user account information, device information, or other related information. This invention will treat the relevant information and its processing with the utmost diligence.
[0185] This invention places great emphasis on the security of relevant information and has adopted reasonable and feasible security protection measures that comply with industry standards to protect user information and prevent unauthorized access, public disclosure, use, modification, damage or loss of relevant information.
[0186] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.
[0187] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.
[0188] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims
1. A method for augmented reality content recommendation and expressive presentation based on user behavior perception, characterized in that, include: Constructing state representations for multiple virtual characters in augmented reality scenarios; The state representation includes the three-dimensional spatial position, facing direction, and occlusion status of each virtual character; A joint objective function is constructed to evaluate the rationality of the layout of virtual characters; the joint objective function includes a spatial cost function and an orientation cost function, wherein the spatial cost function is used to quantify the rationality of the spatial relationship between characters, and the orientation cost function is used to quantify the coordination and diversity of the characters' orientations; Based on the attribute information of the virtual character, adaptive weights are generated through a pre-built weight prediction model, and the adaptive weights are used to adjust the cost terms related to the importance of the character in the spatial cost function. Based on the joint objective function, the position and orientation of all virtual characters are iteratively optimized to obtain an optimized layout; Expressive presentation of the optimized layout in augmented reality scenarios; The process of generating adaptive weights through a pre-built weight prediction model includes: Maintain a weight cache pool, which stores the historical weights corresponding to each virtual character in the historical optimized layout; For each virtual character in the current optimized layout, the historical weight of each virtual character is obtained from the weight cache pool, and the absolute difference between the predicted weight of the current optimized layout and the weight mean of all the historical weights is determined. Based on the absolute difference and a preset sensitivity threshold, an influence factor is assigned to the predicted weight and each of the historical weights. The influence factor includes a first influence factor for the predicted weight and a second influence factor for each of the historical weights. The sum of the first influence factor and all second influence factors is 1. The second influence factor for each historical weight decreases as the storage time increases. If the absolute difference is greater than or equal to the preset sensitivity threshold, the first influence factor is greater than the second sum of all second influence factors. If the absolute difference is less than the preset sensitivity threshold, the first influence factor is less than the second sum of all second influence factors. The fusion weight is obtained by multiplying the predicted weights and each of the historical weights with their respective influence factors, and then summing the products. The fusion weights are used as the adaptive weights, and the adaptive weights are updated in the weight cache pool for smooth updates in subsequent rounds.
2. The augmented reality content recommendation and expressive presentation method based on user behavior perception according to claim 1, characterized in that, Construct a joint objective function for evaluating the rationality of virtual character layout, including: The spatial cost function is constructed as the sum of the collision cost function, the boundary cost function, the center cost function, the distribution cost function, and the region cost function. The collision cost function is calculated based on the intersection-union ratio of the character bounding boxes. The boundary cost function penalizes virtual characters that are close to the scene boundary. The center cost function encourages virtual characters to move closer to the scene center according to their corresponding importance weights. The distribution cost function encourages virtual characters to be evenly distributed in space. The region cost function encourages virtual characters to cover multiple partitioned regions of the scene. The orientation cost function is constructed as the sum of the deviation cost function and the diversity cost function. The deviation cost function calculates the deviation between the current orientation of each virtual character and its ideal orientation, which is determined by the position of the virtual character relative to a preset point of interest. The diversity cost function encourages diversity in the distribution of orientations of all virtual characters.
3. The augmented reality content recommendation and expressive presentation method based on user behavior perception according to claim 1, characterized in that, Before generating adaptive weights using a pre-built weight prediction model, the following steps are also included: The categorical features in the attribute information are embedded and encoded to obtain a categorical feature vector; The numerical features are normalized to obtain the numerical feature vector. The category feature vector and the numerical feature vector are concatenated to obtain a unified feature vector; wherein, the unified feature vector is input into the weight prediction model for weight prediction.
4. The augmented reality content recommendation and expressive presentation method based on user behavior perception according to claim 3, characterized in that, Before maintaining the weight cache pool, the following is also included: The unified feature vector is input into the weight prediction model to predict the weights, thereby obtaining the predicted weights for the current optimized layout.
5. The augmented reality content recommendation and expressive presentation method based on user behavior perception according to claim 1, characterized in that, Based on the joint objective function, the positions and orientations of all virtual characters are iteratively optimized to obtain an optimized layout, including: Using an attention-based neural network model, an initial layout of the global structure is generated based on the semantic features of the current augmented reality scene and the initial state of the virtual character, which serves as the current layout state. Based on the current layout, all virtual characters are divided into multiple spatial function groups according to spatial proximity or functional similarity. In each iteration, a group-level action is sampled from the preset action space and executed for each spatial function group. The group-level action is used to coordinate the position and orientation of all virtual characters in the group in the three-dimensional space, thereby generating an updated layout state. Based on the updated layout state and the joint objective function, a group-level reward is calculated for each spatial function group that has performed a group-level action; the group-level reward is the negative of the sum of the spatial cost function and the orientation cost function corresponding to the spatial function group. An optimization algorithm based on policy gradient is adopted to update the action selection strategy of each spatial function group according to the group-level reward, and the updated layout state is used as the current layout state for the next iteration until the preset convergence condition is met, and the final optimized layout is output.
6. The augmented reality content recommendation and expressive presentation method based on user behavior perception according to claim 1, characterized in that, Before constructing state representations for multiple virtual characters in an augmented reality scene, the following steps are also included: Real-time acquisition of user behavior perception data in augmented reality scenarios; the behavior perception data includes the user's gaze direction, head posture, interaction actions, and dwell time; Based on the behavioral perception data, the user's real-time attention area in the current scene is estimated through a pre-trained visual attention mechanism or a deep neural network model. Calculate the correlation score between each candidate virtual character and the real-time attention area; Candidate virtual characters are sorted according to the correlation score, and multiple virtual characters with scores higher than a preset threshold are selected.
7. The augmented reality content recommendation and expressive presentation method based on user behavior perception according to claim 6, characterized in that, Calculate the relevance score between each candidate virtual character and the real-time attention area, including: The spatial distance between the location of the candidate virtual character and the real-time attention area is obtained, the matching degree between the semantic tags of the candidate virtual character and the user's historical interests is obtained, and the angle between the orientation of the candidate virtual character and the user's line of sight is obtained. A spatial attenuation factor is generated based on the spatial distance, an orientation correction coefficient is generated based on the included angle, and a semantic base score is generated based on the matching degree. By fusing the spatial attenuation factor and the orientation correction coefficient, the real-time attention gain coefficient is obtained; The semantic base score is modulated using the real-time attention gain coefficient to obtain the modulated semantic interest score. The modulated semantic interest score is weighted and combined with the basic visibility score calculated based on spatial distance to output the final relevance score.
8. The augmented reality content recommendation and expressive presentation method based on user behavior perception according to claim 7, characterized in that, Generating a spatial attenuation factor based on the spatial distance includes: According to a preset attenuation function, the spatial distance is mapped to a value between zero and one as the spatial attenuation factor; wherein, the closer the distance, the higher the mapped spatial attenuation factor value. Based on the included angle, an orientation correction coefficient is generated, including: According to the preset mapping rules, the included angle is mapped to a value between zero and one as the orientation correction coefficient; wherein, the smaller the included angle, the more consistent the character's orientation is with the user's line of sight, and the higher the orientation correction coefficient value obtained by mapping. Based on the matching degree, a semantic base score is generated, including: Based on the preset scoring rules, an initial semantic base score is assigned to the candidate virtual character according to the degree of matching between the semantic tag and the user's historical interests; the higher the degree of matching, the higher the assigned semantic base score.