A method and device for Peking opera interactive experience based on motion capture

Abstract

This application relates to a method and apparatus for interactive Peking Opera experience based on motion capture, belonging to the field of interactive experience technology. The method includes: acquiring user action data from multiple general users using motion capture; calculating group attention parameters and group learning difficulty parameters for multiple target stylized Peking Opera movements; extracting high-value movement slices from the multiple target stylized Peking Opera movements based on the group attention parameters and group learning difficulty parameters to obtain a movement slice sequence; and dynamically adjusting the experience content of subsequent Peking Opera interactive experiences based on the movement slice sequence. This invention solves the problem that traditional Peking Opera interactive experiences often push fixed content, failing to adapt to the actual interests and learning difficulties of ordinary users, resulting in insufficient exposure to content of interest and a lack of targeted guidance for difficult movements, thereby reducing experience participation and learning effectiveness.

Description

Technical Field

[0001] This application relates to the field of interactive experience, and in particular to a method and device for interactive experience of Peking Opera based on motion capture. Background Technology

[0002] The promotion of Peking Opera culture today needs to cater to the public's experience needs, and interactive experiences have become an important way to bridge the gap between ordinary users and the stylized movements of Peking Opera.

[0003] However, traditional interactive experiences of Peking Opera often adopt a fixed content push model, failing to adapt to the actual interests and learning difficulties of ordinary users. This not only results in insufficient exposure to content that users are interested in, but also a lack of targeted guidance for movements that are difficult to master, ultimately reducing user engagement and learning effectiveness, which is not conducive to the effective popularization and public acceptance of Peking Opera culture. Summary of the Invention

[0004] This application provides a method and device for interactive Peking Opera experience based on motion capture, which improves the current situation where traditional Peking Opera interactive experiences mostly push fixed content without adapting to the interests and learning difficulties of ordinary users, resulting in insufficient exposure to content that users are interested in and a lack of targeted guidance for mastering movements.

[0005] The embodiments of this application disclose the following technical solutions:

[0006] In a first aspect, embodiments of this application provide a method for interactive Peking Opera experience based on motion capture, the method comprising:

[0007] By combining motion capture, user action data of multiple general users are obtained, wherein the user action data includes limb movement sequences and behavioral action sets;

[0008] Based on the user action data, calculate the group attention parameter and group learning difficulty parameter for multiple target Peking Opera stylized actions;

[0009] Based on the group attention parameter and the group learning difficulty parameter, high-value action slices are extracted from multiple target Peking Opera stylized actions to obtain action slice sequences.

[0010] Based on the aforementioned action slice sequence, the content of the subsequent Peking Opera interactive experience is dynamically adjusted.

[0011] Secondly, embodiments of this application provide a motion capture-based interactive experience device for Peking Opera, the device comprising:

[0012] The user action data acquisition module is used to acquire user action data from multiple general users by combining motion capture. The user action data includes a sequence of limb movements and a set of behavioral actions.

[0013] The group dual-parameter calculation module is used to calculate the group attention parameter and group learning difficulty parameter of multiple target Peking Opera stylized movements based on the user action data.

[0014] The high-value action slice extraction module is used to extract high-value action slices from multiple target Peking Opera stylized actions based on the group attention parameter and the group learning difficulty parameter, and obtain an action slice sequence.

[0015] The experience content adjustment module is used to dynamically adjust the experience content of subsequent Peking Opera interactive experiences based on the action slice sequence.

[0016] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0017] This application proposes a method and device for interactive Peking Opera experience based on motion capture. By acquiring user motion data step-by-step, calculating key parameters, extracting high-value motion slices, and dynamically adjusting the experience content, it achieves precise adaptation of the interactive Peking Opera experience content to the interests, preferences, and learning needs of general users. First, using motion capture technology, the first motion data and first operation data of users within a table are obtained from historical Peking Opera interactive experience logs. Simultaneously, second motion data and second operation data of users outside the table are obtained based on big data analysis. After processing, a sequence of limb movements and a set of behavioral actions for general users are obtained. Then, based on the aforementioned user motion data, the group attention parameters and group learning difficulty parameters for multiple target stylized Peking Opera movements are calculated. Subsequently, these two types of parameters are compared with corresponding preset thresholds to extract high-attractiveness slices and high-difficulty teaching slices, forming a sequence of motion slices. Finally, the experience content is dynamically adjusted based on the motion slice sequence. High-attractiveness slices are serialized according to the group attention parameters and displayed at the top, while associated auxiliary teaching resource packages are generated for high-difficulty teaching slices and can be called upon as needed.

[0018] The technical solution proposed in this application solves the problems of traditional Peking Opera interactive experiences, which mostly push fixed content, cannot adapt to user interests and learning obstacles, and result in low user participation and poor learning effects. It avoids the situation where users do not have enough access to content that is of interest to them and lack guidance for difficult movements due to the lack of targeted content. It improves the personalization and effectiveness of Peking Opera interactive experiences, and at the same time provides technical support that is more in line with the needs of the public for the popularization and dissemination of Peking Opera culture. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 A flowchart illustrating a motion capture-based interactive experience method for Peking Opera provided in this application embodiment;

[0021] Figure 2 This is a schematic diagram of the structure of a motion capture-based interactive experience device for Peking Opera provided in an embodiment of this application.

[0022] The components represented by each number in the attached diagram are explained below:

[0023] User action data acquisition module 01, group dual parameter calculation module 02, high-value action slice extraction module 03, experience content adjustment module 04. Detailed Implementation

[0024] This application provides a motion capture-based interactive experience method and device for Peking Opera to address the technical problem that existing interactive experiences of Peking Opera often involve fixed content push, failing to adapt to the actual interests and learning difficulties of ordinary users, resulting in insufficient exposure to content of interest and a lack of targeted guidance for difficult movements, thereby reducing user engagement and learning effectiveness.

[0025] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0026] In the description of this application, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the stated features. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0027] In the description of this application, the term "for example" is used to mean "used as an example, illustration, or description." Any embodiment described as "for example" in this application is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is provided to enable any person skilled in the art to make and use the invention. Details are set forth in the following description for purposes of explanation. It should be understood that those skilled in the art will recognize that the invention can be made without using these specific details. In other instances, well-known structures and processes will not be described in detail to avoid obscuring the description of the invention with unnecessary detail. Therefore, the invention is not intended to be limited to the embodiments shown, but is consistent with the broadest scope of the principles and features disclosed in this application.

[0028] Example 1, as shown in the appendix Figure 1 As shown, this application provides a method for interactive Peking Opera experience based on motion capture, the method including the following steps:

[0029] S110: Combine motion capture to obtain user action data from multiple general users, wherein the user action data includes a sequence of limb movements and a set of behavioral actions;

[0030] In this embodiment of the application, in the scenario of providing interactive Peking Opera experiences to a broad non-professional audience, in order to comprehensively collect information that reflects users' real action performance and interaction preferences, it is necessary to obtain relevant user data from different channels and perform targeted processing to ensure the accuracy and effectiveness of subsequent extraction of high-value action slices and optimization of experience content.

[0031] Specifically, based on historical Peking Opera interactive experience logs, the first action data and first operation data generated by users in the past experience process are filtered and extracted from the logs. These data directly reflect the action characteristics and operation habits of existing users.

[0032] At the same time, based on big data, we extensively collect second action data and second operation data from users outside the table, covering potential user groups who have not formed historical log records, so as to ensure the breadth and representativeness of the data sample.

[0033] Furthermore, the acquired first and second action data are processed uniformly using motion capture technology. By accurately collecting and integrating information such as the user's limb movement trajectory and posture changes, a sequence of limb movements of general users is formed to fully present the actual situation of the user imitating the stylized movements of Peking Opera.

[0034] Meanwhile, using the target Peking Opera stylized movements as an index, statistical analysis is performed on the first and second operation data to sort out the user's operational behavior characteristics in different target action-related interactions, and then integrate them to form a set of behavioral actions to comprehensively cover the user's operational feedback information in the process of learning and experiencing the movements.

[0035] This step involves acquiring user data through multiple channels and processing it in different dimensions to form a sequence of body movements and a set of behavioral actions. This provides a complete data foundation for subsequent calculations of group attention parameters and group learning difficulty parameters, ensuring that subsequent adjustments to the user experience are more in line with actual user needs.

[0036] Step S110 in the method provided in this application embodiment includes:

[0037] Based on historical Peking Opera interactive experience logs, obtain the first action data and first operation data of the users in the table;

[0038] Based on big data, acquire the second action data and second operation data of users experiencing the off-table experience;

[0039] Motion capture is performed on the first motion data and the second motion data to obtain the body motion sequence of the general user.

[0040] Using the target stylized Peking Opera movements as an index, statistical analysis is performed on the first operation data and the second operation data to obtain the set of behavioral movements.

[0041] In this embodiment of the application, in order to comprehensively collect information that reflects the action performance and interaction preferences of general users in the interactive experience of Peking Opera, it is necessary to obtain action and operation-related data from different user groups, and form complete user action data through targeted processing, so as to ensure the accuracy and effectiveness of subsequent extraction of high-value action slices and optimization of experience content.

[0042] Specifically, based on historical Peking Opera interactive experience logs, the first action data and first operation data of the users in the table are obtained.

[0043] Among them, the historical Peking Opera interactive experience logs need to cover user experience records for a certain period of time, such as log data from the past 6-12 months. This includes user experience behavior during different seasons and holidays, as well as behavioral changes of users who have participated in the experience for a long time, so as to ensure that the first action data and first operation data obtained can fully reflect the real experience status of users in the table.

[0044] Meanwhile, during the data acquisition process, it is necessary to ensure the accuracy of the action data and the integrity of the operation data recorded in the historical logs. For example, the action data should include details such as the angle of the limb joints and the movement trajectory when the user imitates the stylized movements of Peking Opera, and the operation data should fully record the various interactive operations of the user during the experience.

[0045] Furthermore, based on big data technology, second action data and second operation data of off-balance-sheet users are obtained. Off-balance-sheet users include potential users who have not yet generated historical experience logs, such as visitors experiencing Peking Opera interactive experiences for the first time, and users who only browse related content on mobile devices without engaging in in-depth experiences.

[0046] During the acquisition process, it is necessary to use legal and compliant big data collection channels to ensure the security of the data source and the confidentiality of user information. At the same time, it is necessary to ensure that the format and dimensions of the second action data and the second operation data are consistent with the first action data and the first operation data of the user in the table, so as to facilitate unified processing in the future.

[0047] Furthermore, after acquiring the first action data and the second action data, motion capture processing is performed on the two types of data to obtain the body movement sequence of general users.

[0048] Specifically, during motion capture, visual image data of the user's limb movements are collected using image sensors, and the collected visual image data is processed using motion recognition and analysis technology to accurately extract and integrate the user's limb movement information when imitating stylized movements in Peking Opera.

[0049] For example, the system captures the arm swing trajectory, leg movement path, and body posture changes when users imitate movements such as cloud hands and stage steps, transforming scattered motion data into a continuous and complete sequence of limb movements to fully present the actual situation of general users imitating stylized movements of Peking Opera.

[0050] At the same time, consistency verification is required for the captured limb movement sequences to ensure the uniformity of movement data from different users in dimensions such as time axis and spatial coordinates.

[0051] Finally, using the target Peking Opera stylized movements as an index, statistical analysis is performed on the first and second operation data to obtain a set of behavioral actions.

[0052] In the method provided in this application embodiment, the set of behavioral actions includes at least the number of clicks, single playback duration, and number of loop playbacks corresponding to multiple target Peking Opera stylized action videos.

[0053] Specifically, in the statistical analysis process, for each target stylized Peking Opera movement, it is necessary to count the number of clicks on the corresponding video by users both inside and outside the table, for example, to count the total number of clicks on the video of the "Lying Fish" movement; to record the single playback duration of the video of the movement, for example, the average duration of the video of the "Starting the Show" movement watched by users in its entirety; and to calculate the number of times the video of the movement is looped, for example, the number of times the user repeatedly watches the video of the "Feather Skill" movement.

[0054] At the same time, it is necessary to deduplicatize and integrate the statistical data, such as excluding invalid data generated by accidental clicks, merging the operation records of the same user on different devices for the same action video, and finally forming a set of behavioral actions that can clearly reflect the user's preference for different target Peking Opera stylized action interactions.

[0055] Ultimately, through the above multi-step data acquisition and processing, complete user action data is formed, providing reliable data support for subsequent steps.

[0056] S120: Based on the user action data, calculate the group attention parameter and group learning difficulty parameter of multiple target Peking Opera stylized actions;

[0057] In this embodiment of the application, after acquiring general user action data through motion capture, in order to accurately identify users' interest preferences and learning difficulties in different target Peking Opera stylized actions, it is necessary to calculate the group attention parameter and the group learning difficulty parameter based on the collected body action sequences and behavioral action sets, so as to ensure that high-value action slices can be extracted in a targeted manner and provide data support for dynamically adjusting the experience content.

[0058] In the method provided in this application embodiment, the group attention parameter is determined based on the interaction concentration index of general users towards the target Peking Opera stylized actions, including:

[0059] Based on the aforementioned body movement sequence, statistical analysis was performed to obtain the cumulative number of times that general users selected the target Peking Opera stylized movements;

[0060] Based on the aforementioned body movement sequence, the average single dwell time of general experience users in the imitation and practice of the target Peking Opera stylized movements is calculated;

[0061] Based on the set of behavioral actions, statistical analysis was conducted to obtain the percentage of click actions by general users on the target stylized Peking Opera actions.

[0062] Based on the set of behavioral actions, statistical analysis was conducted to obtain the typical action replay ratio of general experience users on the target stylized Peking Opera actions.

[0063] The interaction concentration index is obtained by weighting the cumulative number of times, the average single dwell time, the percentage of click actions, and the typical action replay ratio.

[0064] Specifically, the first step is to statistically analyze the cumulative number of times general users selected the target stylized Peking Opera movements based on body movement sequences. The body movement sequences comprehensively record the types of movements selected and imitated by users during the experience, and the cumulative count must cover the movement selection records of all general users within a preset period.

[0065] For example, within a one-month statistical period, the total number of times all users selected different movements such as cloud hands, platform steps, and lying fish was recorded. If the cumulative number of selections for the Lingzi Gong movement reached 500 times, while the cumulative number of selections for the Qiba movement was 200 times, it indicates that the initial selection intention for Lingzi Gong was stronger among users, and it is a potentially high-attention movement.

[0066] Secondly, the average single-session dwell time of general experience users in the target Peking Opera stylized movement imitation practice session is calculated based on the body movement sequence. Dwell time directly reflects the user's engagement in the movement practice; the calculation requires first extracting the single-session dwell time of each user practicing a single target movement, and then calculating the average of all users.

[0067] For example, when users practiced the large knife flourish, user A spent 15 minutes, user B spent 12 minutes, and user C spent 18 minutes. The average time spent on this action was calculated to be 15 minutes. If the average time spent on the circular motion was only 8 minutes, it would indicate that users spent more time practicing the large knife flourish and had a stronger willingness to maintain focus on the action.

[0068] Furthermore, based on the statistical analysis of behavioral action sets, the percentage of clicks made by general users on the target Peking Opera stylized actions is analyzed. The behavioral action set includes the user's click records on videos related to each target action. The percentage of clicks is calculated by determining the proportion of clicks on a single target action to the total number of clicks on all user action videos.

[0069] For example, if a user clicks on various action videos a total of 100 times during the experience, and clicks on the water sleeve exercise video 35 times, then the click rate for the water sleeve exercise is 35%. If the click rate for the walking side exercise is only 10%, it indicates that the user has a higher willingness to actively click on the water sleeve exercise, further confirming the potential for attention to this action.

[0070] Furthermore, based on the statistical analysis of behavioral action sets, the typical action replay rate of general experience users on the target Peking Opera stylized actions is analyzed. The typical action replay rate needs to be calculated as the ratio of the number of times users replay key segments in the target action to the total number of times the action video is played. Key segments are usually the parts of the action that highlight technical points or have vivid visual effects.

[0071] For example, if a user watches the video of the kicking gun action 5 times, and rewatches the segment of turning around with the gun 12 times, then the typical rewatch rate of the kicking gun action is 2.4 (12 / 5=2.4); if the typical rewatch rate of the hammer-wielding action is 1.2, it means that the typical segments of the kicking gun action are more attractive to users to watch repeatedly, which is an important reflection of user attention.

[0072] Finally, the interaction concentration index is obtained by weighting the cumulative number of times, average single dwell time, click action ratio, and typical action replay ratio.

[0073] Specifically, when calculating the weighted average, reasonable weights should be set according to the degree of influence of each dimension on attention. For example, the cumulative number of times and the percentage of clicks reflect the user's willingness to actively choose, and the weights can be set to 30% respectively. The average single dwell time reflects the user's continuous engagement, and the weight is set to 25%. The typical action replay ratio reflects the user's depth of interest in the details of the action, and the weight is set to 15%.

[0074] For example, taking the Lingzi Gong (a type of martial arts exercise) as an example, if its cumulative number of times scores 80 points, average single dwell time scores 75 points, click action ratio scores 85 points, and typical action replay rate scores 90 points, the interaction concentration index can be calculated as (80×30%+75×25%+85×30%+90×15%)=81.75 points after weighted calculation. This score is the indicator for measuring the public's attention to the Lingzi Gong exercise.

[0075] In the method provided in this application embodiment, the group learning difficulty parameter is determined based on the convergence of general experience users imitating the target Peking Opera stylized movements, including:

[0076] Based on the body movement sequence, extract the deviation vector between each general experience user and the standard movement template when imitating the target Peking Opera stylized movement;

[0077] Cluster analysis is performed on the deviation vectors of multiple general experience users, and the silhouette coefficient of the clustering results is calculated;

[0078] The variance of multiple deviation vectors is statistically analyzed to obtain the variance variation trend over time.

[0079] The convergence of the action is calculated by weighting the profile coefficient and the variance variation trend.

[0080] Specifically, the deviation vector between each general experience user and the standard action template when imitating the target Peking Opera stylized actions is first extracted based on the body movement sequence.

[0081] Among them, the limb movement sequence fully records the details such as the joint angles and movement trajectories of the limbs when the user imitates the movement, while the standard movement template is a movement benchmark established based on the professional norms of Peking Opera, including the standard joint angle range, standard movement path, key posture nodes and other content of the target movement.

[0082] When extracting the deviation vector, each key dimension of the user's action needs to be compared with the corresponding dimension of the standard template. The difference value is calculated and integrated to form a vector. For example, for the cloud hand action, the standard template stipulates that the maximum bending angle of the elbow joint when the arms swing is 120° and the movement trajectory of the hands must maintain a horizontal arc. If a user imitates it and the maximum bending angle of the elbow joint is 90° and the movement trajectory of the hands is offset downward by 5cm, then these two differences and other key dimension differences are integrated to form the deviation vector of the user's imitation of the cloud hand action, so as to intuitively reflect the degree of deviation between the individual action and the standard template.

[0083] Furthermore, cluster analysis is performed on the deviation vectors of multiple general users, and the silhouette coefficient of the clustering results is calculated. Specifically, the cluster analysis grouped all users' deviation vectors according to similarity, with deviation vectors of higher similarity belonging to the same cluster to represent the distribution of deviations within the group.

[0084] In addition, the silhouette coefficient is used to evaluate the clustering effect. Its value ranges from -1 to 1. The closer the silhouette coefficient is to 1, the higher the similarity of the deviation vectors within the same cluster and the greater the difference in the deviation vectors between different clusters. This means that the action deviations of the user group are significantly differentiated, with some users having small deviations and others having large deviations. This reflects that the learning difficulty of the action varies greatly among different users, and the overall learning threshold is relatively high.

[0085] Conversely, a silhouette coefficient closer to -1 indicates poor clustering, a chaotic distribution of user biases, or that the difficulty of the action is too high, resulting in most users having extremely large biases and making it difficult to form effective clusters. A silhouette coefficient close to 0 indicates that the distribution of user biases is relatively uniform, and the difficulty of the action is moderate for the group.

[0086] For example, after clustering the deviation vectors of 50 users for the "lying fish" action, the silhouette coefficient was -0.2. This indicates that the deviation vectors of most users have small differences, but the overall deviation from the standard template is far. This suggests that the "lying fish" action is difficult to learn for this group, and most users have difficulty mastering the standard action.

[0087] Furthermore, the variance of multiple deviation vectors was statistically analyzed to determine its variation over time. Variance reflects the dispersion of the user group's deviation vectors; a larger variance indicates greater differences in action deviations among users, leading to less stable overall imitation performance. Conversely, a smaller variance indicates smaller differences in deviations among users, resulting in more uniform imitation performance.

[0088] When analyzing the trend of variance, time should be plotted on the horizontal axis and variance on the vertical axis to observe how the variance changes over time. For example, for a catwalk movement, the variance of the deviation vector was 80 in the first week, dropped to 60 in the second week, and further decreased to 45 in the third week, showing a continuous downward trend. This indicates that as users practice for longer periods, the dispersion of movement deviation within the group gradually decreases, and the users' imitation effects gradually become more uniform. The learning difficulty of this movement is relatively low, and the user group can gradually master it through practice.

[0089] However, when it comes to the Lingzi Gong movements, the variance of the deviation vectors for three consecutive weeks were 120, 115, and 118, respectively, remaining at a high level without any obvious downward trend. This indicates that even after practice, the deviations in the movements of the user group are still large and difficult to reduce, reflecting that the Lingzi Gong movements are difficult to learn and that the user group cannot master the standard movements in a short period of time.

[0090] Finally, the motion convergence is calculated by weighting the silhouette coefficient and the variance trend. The weighting calculation requires setting weights based on the degree of influence of each indicator on the motion convergence. For example, the silhouette coefficient reflects the stability of the group deviation distribution, with a weight of 40%; the variance trend reflects the improvement of the group deviation over time, with a weight of 60%. The specific weights can be adjusted according to the emphasis on difficulty judgment in the actual experience scenario.

[0091] During the calculation process, the original data of the profile coefficient and variance change trend need to be standardized first, and then the weighted total score is calculated according to the weight. This weighted total score is the action convergence.

[0092] Among them, the higher the action convergence score, the better the profile coefficient and the more obvious the variance decrease trend. The user group's actions are more likely to converge to the standard template, and the lower the difficulty of action learning. The lower the action convergence score, the worse the profile coefficient and the no decrease or even increase in variance. The user group's actions are difficult to converge, and the higher the difficulty of action learning.

[0093] For example, the standardized profile coefficient score of the platform step movement is 70 points and the variance trend score is 85 points. After weighted calculation, the movement convergence is (70×40%+85×60%)=79 points, indicating that its learning difficulty is relatively low. The standardized profile coefficient score of the feather exercise is 30 points and the variance trend score is 25 points. After weighted calculation, the movement convergence is (30×40%+25×60%)=27 points, indicating that its learning difficulty is relatively high.

[0094] Ultimately, by using the weighted calculation of multi-dimensional user action data to obtain the group attention parameters and group learning difficulty parameters, we can accurately locate the points of interest and learning difficulties of general users in the target Peking Opera stylized movements, providing data basis for extracting high-attractive and high-difficulty teaching segments from multiple target movements.

[0095] S130: Based on the group attention parameter and the group learning difficulty parameter, extract high-value action slices from multiple target Peking Opera stylized actions to obtain an action slice sequence;

[0096] In this embodiment of the application, in order to accurately select high-value action content that meets the user's interests and learning needs, and to avoid the lack of targeted experience content due to indiscriminate treatment of all actions, it is necessary to classify and mark actions by comparing preset thresholds to form an orderly sequence of action slices, which will provide guidance for the subsequent dynamic adjustment of Peking Opera interactive experience content.

[0097] First, the group attention parameter of each target Peking Opera stylized movement needs to be compared with the first preset threshold, and its group learning difficulty parameter needs to be compared with the second preset threshold.

[0098] The first and second preset thresholds are confidence thresholds derived from statistical analysis of historical interaction data, which can objectively reflect the general level of attention and average learning ability of the general user group to the stylized movements of Peking Opera.

[0099] Specifically, when the group attention parameter of a certain stylized Peking Opera movement is higher than the first preset threshold, it indicates that the movement has a high level of interest and appeal among users, and it is marked as a high-attractiveness segment; when the group learning difficulty parameter of a certain stylized Peking Opera movement is higher than the second preset threshold, it indicates that the movement has a high learning threshold and many learning obstacles for users, and it is marked as a high-difficulty teaching segment.

[0100] Finally, after classifying and labeling all target actions, highly engaging slices and challenging instructional slices are integrated to form a sequence of action slices containing both types of high-value content. This sequence covers both actions that users are most interested in and actions that users most need guidance on. Through the extraction methods described above, it is ensured that subsequent adjustments to the user experience content accurately match the user's core needs, avoiding the waste of experience resources by ineffective content.

[0101] Step S130 in the method provided in this application embodiment includes:

[0102] The group attention parameter is compared with a first preset threshold, and the group learning difficulty parameter is compared with a second preset threshold.

[0103] When the group attention parameter is higher than the first preset threshold, the corresponding target Peking Opera stylized action is marked as a high-attractive slice;

[0104] When the group learning difficulty parameter is higher than the second preset threshold, the corresponding target Peking Opera stylized movement is marked as a high-difficulty teaching segment;

[0105] Wherein, the first preset threshold and the second preset threshold are confidence thresholds based on statistical analysis.

[0106] In this embodiment of the application, in order to avoid the lack of specificity of the experience content due to the indiscriminate processing of all actions, it is necessary to classify the action slices by comparing preset thresholds and parameters to form clear high-value action slice types, so as to provide a clear basis for subsequent integration of action slice sequences and dynamic adjustment of experience content.

[0107] Specifically, the first step is to compare the group attention parameter with the first preset threshold, and the group learning difficulty parameter with the second preset threshold.

[0108] The first and second preset thresholds are both confidence thresholds based on statistical analysis. During the setting process, it is necessary to collect interaction data between general users and the stylized movements of each target Peking Opera over a relatively long period of time, including historical group attention parameter values ​​and group learning difficulty parameter values. The confidence interval that can reflect the general characteristics of the group is calculated through data statistical methods, and the threshold size is finally determined.

[0109] For example, when setting the first preset threshold, the group attention parameters of all target actions in the past year are statistically analyzed, and the upper limit of the 95% confidence interval is calculated as the first preset threshold to ensure that the first preset threshold can effectively distinguish actions that exceed the general level of group attention.

[0110] Similarly, when setting the second preset threshold, the group learning difficulty parameters of all target actions in the past year are also statistically analyzed, and the upper limit of the 95% confidence interval is used as the second preset threshold to ensure that the second preset threshold can accurately identify high-difficulty actions that exceed the average learning ability of the group.

[0111] Furthermore, after setting the threshold, the group attention parameter is compared with the first preset threshold. When the group attention parameter of a certain target Peking Opera stylized movement is higher than the first preset threshold, it indicates that the movement has a significantly higher level of attention among users than most movements. Whether it is the number of times users select it, the duration of practice, the click-through rate, or the replay rate, it is at a high level, and it is marked as a high-attractive slice.

[0112] Furthermore, the group learning difficulty parameter is compared with the second preset threshold. When the group learning difficulty parameter of a certain target Peking Opera stylized movement is higher than the second preset threshold, it indicates that the learning threshold of this movement is significantly higher than that of most movements. During the imitation process, the deviation vector between the user group and the standard movement template is large, and the convergence speed of the deviation is slow. Most users find it difficult to master quickly, and this is marked as a high-difficulty teaching segment.

[0113] During the comparison and labeling process, it is also necessary to ensure that each target Peking Opera stylized movement completes the comparison of two types of parameters with the corresponding thresholds, and that the labeling results are accurate. Some movements may only meet the conditions for high-attractiveness slices, some movements may only meet the conditions for high-difficulty teaching slices, and there may be a few movements that are labeled as both slice types because both types of parameters are higher than the corresponding thresholds. In this case, neither label needs to be removed, and the movements should be retained in their entirety so that their high-attractiveness and high-difficulty characteristics can be taken into account in subsequent adjustments to the experience content.

[0114] Ultimately, through the threshold comparison and labeling operations described above, numerous target Peking Opera stylized movements were clearly divided into two categories of high-value segments: highly attractive segments and highly difficult teaching segments. This laid the foundation for subsequent integration of movement segment sequences and targeted adjustments to the experience content, thereby improving the effectiveness of the Peking Opera interactive experience.

[0115] S140: Based on the action slice sequence, dynamically adjust the experience content of subsequent Peking Opera interactive experiences.

[0116] In this embodiment of the application, in order to make the subsequent interactive Peking Opera experience content more in line with the interests and learning needs of general users, it is necessary to adopt differentiated adjustment strategies for different types of high-value action slices to improve the relevance and effectiveness of the experience content, so that users can better participate in the Peking Opera action experience and learning.

[0117] Specifically, adjustments are first made to the highly attractive slices. These slices need to be serialized based on group attention parameters, and then sorted in descending order of group attention parameters to form an ordered display sequence.

[0118] In subsequent interactive Peking Opera experience scenarios, based on the above serialization results, highly attractive slices are displayed at the top, allowing users to prioritize access to the most attention-grabbing action content, reducing the time cost for users to find content of interest, and enhancing users' interest in the experience content.

[0119] Furthermore, adjustments will be made to the high-difficulty teaching segments. Big data technology will be used to generate and associate at least one supplementary teaching resource package for each high-difficulty teaching segment. This supplementary resource package must contain content that helps users understand and master high-difficulty movements, including at least animations breaking down the movement trajectory, typical error prompts, and explanations of key pose freeze-frames. These resources will help users clearly understand the complete process of the movement, common errors, and key pose requirements.

[0120] At the same time, a mechanism is established to link supplementary teaching resource packages with high-difficulty teaching segments. When a user selects a high-difficulty teaching segment as the content to experience, the associated supplementary teaching resource package can be automatically invoked to provide the user with timely teaching support.

[0121] This step, by prioritizing the display of highly engaging segments and configuring supplementary teaching resource packages for high-difficulty teaching segments, ensures that subsequent interactive Peking Opera content prioritizes meeting users' interests and needs, effectively improving the adaptability of the content and providing users with a higher-quality interactive experience.

[0122] Step S140 in the method provided in this application embodiment includes:

[0123] Based on the group attention parameters, the highly attractive slices are serialized, and the highly attractive slices are displayed at the top in the Peking Opera interactive experience scene according to the serialization results;

[0124] Based on big data, at least one auxiliary teaching resource package is generated and associated for each of the high-difficulty teaching segments. The auxiliary teaching resource package includes at least motion trajectory decomposition animation, typical error prompts and key posture freeze-frame explanations.

[0125] When the advanced teaching segment is selected as the experience content, the auxiliary teaching resource package is invoked.

[0126] In this embodiment of the application, in order to avoid low user participation and poor learning results due to fixed and untargeted experience content, it is necessary to adopt appropriate adjustment strategies for the two types of high-value segments to improve the practicality of the experience content and the user experience, and help the effective dissemination of Peking Opera stylized movements and efficient user learning.

[0127] Specifically, adjustments are first made to highly attractive segments, namely, serialization and pinning based on group attention parameters. During serialization, the group attention parameter is used as the sole sorting criterion. This parameter is derived by weighting multiple dimensions of data, including the cumulative number of times users selected the target action, the average single-session dwell time during imitation practice, the percentage of action clicks, and the replay rate of typical actions. It objectively reflects the level of interest of general users in different highly attractive segments; the higher the parameter value, the stronger the user interest.

[0128] For example, in a batch of highly attractive slices, the group attention parameter of the cloud hand movement is 90, the group attention parameter of the stage step movement is 82, the group attention parameter of the reclining fish movement is 75, and the group attention parameter of the circular movement is 68. Then, the sequence is completed in the order of cloud hand - stage step - reclining fish - circular movement to form an ordered sequence of highly attractive slices.

[0129] Furthermore, after the serialization process is completed, the most attractive slices need to be displayed at the top of the screen in the Peking Opera interactive experience scene based on the sequence results. The interface design of the experience scene needs to dedicate a specific area for this top-display feature, located at the user's visual focus, such as a popular recommendation bar at the top of the screen. The most attractive slices, ranked first after serialization, will be presented sequentially in this top-display area.

[0130] For example, the above-mentioned cloud hand, stage walk, and reclining fish slices are prioritized and placed in the top three of the popular recommendations. After entering the experience interface, users can directly see the most popular action content without scrolling the interface or searching for menus, reducing the operational cost for users to obtain content of interest. At the same time, by prioritizing the display of highly interesting content, users' attention can be quickly attracted, and users' initial willingness to participate can be increased.

[0131] Furthermore, adjustments are made to high-difficulty teaching segments, with a focus on generating and associating supplementary teaching resource packages with big data, while enabling on-demand access to these resource packages.

[0132] Specifically, when generating supplementary teaching resource packages, it is necessary to rely on big data technology to extensively collect common problems and typical error cases of general users when learning various high-difficulty Peking Opera stylized movements, as well as the core explanation points and technical breakdown logic of such movements in professional Peking Opera teaching, and customize exclusive supplementary teaching resource packages for each high-difficulty teaching segment.

[0133] The supplementary teaching resource package must include at least three core types of content. The first is motion trajectory decomposition animation, which breaks down high-difficulty movements into multiple consecutive keyframes along the timeline, clearly presenting the complete path of limb movement. For example, for the high-difficulty movement of feathering, the movements of circling the feather, picking the feather, and shaking the feather are decomposed frame by frame, and the direction and amplitude of the movement of the head, neck, and arms in each frame are marked.

[0134] Secondly, there are typical error prompts, which summarize the common mistakes that most users make when imitating the action. For example, common problems in the big knife flower action include insufficient arm swing range, incorrect wrist rotation angle, and unstable body center of gravity. The error characteristics and their impact on the continuity of subsequent actions are marked in the form of text and diagrams.

[0135] Finally, there is the explanation of key posture freezes, which involves capturing the core postures during the execution of the action and freezing them, such as the postures of raising the armor and pressing the sword in the opening move. The standard angles of the shoulder, elbow, and knee joints in this posture, as well as the power points of the waist and legs, are marked to help users understand the key details of the standard movement.

[0136] Furthermore, after generating the supplementary teaching resource package, a unique association mechanism between the resource package and the high-difficulty teaching slice needs to be established. Each supplementary teaching resource package is bound and stored with its corresponding high-difficulty teaching slice to ensure that the two form a one-to-one mapping relationship.

[0137] Specifically, when a user selects a high-difficulty teaching segment as the current experience content during the interactive experience of Peking Opera, a resource retrieval command will be automatically triggered. Based on the preset association, the auxiliary teaching resource package bound to the high-difficulty teaching segment will be quickly retrieved and displayed simultaneously on the side or bottom area of ​​the experience interface for the user to view at any time.

[0138] For example, when a user selects the advanced instruction segment of kicking a gun, an auxiliary teaching resource package will be immediately displayed, which includes an animation of the kicking motion trajectory breakdown, common error prompts, and explanations of key pose freezes. During the imitation practice, the user can click to view the contents of the resource package at any time, compare the differences between their own movements and the standard movements, correct errors in a timely manner, and improve the learning effect.

[0139] Throughout the implementation process, the adjustment effects also need to be dynamically verified and optimized. For the serialization and top display of highly attractive slices, it is necessary to regularly collect statistics on the user's selection rate, dwell time, and completion rate of the top slices. If the selection rate of a certain top slice is lower than expected for several consecutive periods, it is necessary to re-examine the calculation process of its group attention parameters to confirm whether there are any data statistical biases or improper parameter weight settings that cause the sorting results to be inconsistent with the user's actual interests.

[0140] In addition, for supplementary teaching resource packages, it is also necessary to collect user feedback data after use, including the comprehensibility of the resource package content and the degree of help in correcting errors. If most users report that the action trajectory decomposition of a certain resource package is not clear enough or the explanation of key postures is not detailed enough, it is necessary to combine big data to supplement more user learning cases and optimize the resource package content, such as adding multi-angle demonstration videos, to ensure that the supplementary teaching resources can effectively solve the learning difficulties of users.

[0141] Ultimately, by serializing and prioritizing highly engaging segments and configuring and on-demanding auxiliary resource packages for challenging teaching segments, subsequent interactive Peking Opera experiences can not only prioritize meeting users' interests and needs but also address their learning obstacles in a targeted manner. This effectively improves the adaptability and practicality of the experience content, helping users participate in Peking Opera action experiences and learning more efficiently.

[0142] The embodiments of this application, through the specific implementation methods described above, achieve the following technical effects:

[0143] This application proposes a motion capture-based interactive experience method for Peking Opera. First, combining motion capture technology with multi-channel data collection, it obtains the first action and first operation data of users within a table from historical Peking Opera interactive experience logs. Simultaneously, it collects the second action and second operation data of users outside the table based on big data. Visual images of body movements are acquired using image sensors, and the data is processed using motion recognition and analysis technology to form a sequence of body movements for general users. The operation data is then statistically analyzed using the target Peking Opera stylized movements as an index to obtain a set of behavioral actions containing information such as click count and playback duration. Next, based on the body movement sequence and the set of behavioral actions, a group attention parameter and a group learning difficulty parameter are calculated, respectively. Subsequently, these two parameters are compared with a confidence threshold based on historical data statistics to mark high-attractiveness slices and high-difficulty teaching slices, integrating them to form an action slice sequence. Finally, the experience content is dynamically adjusted based on the sequence. High-attractiveness slices are sequenced according to attention and displayed at the top, while auxiliary teaching resource packages containing action breakdown animations, error prompts, etc., are generated for high-difficulty teaching slices. These resource packages are automatically invoked when the user selects a slice.

[0144] The method provided in this application, through the technical solution of "multi-channel data collection - precise calculation of dual parameters - extraction of high-value slices - adjustment of differentiated content", solves the problem that fixed content push in traditional Peking Opera interactive experience cannot adapt to the interests and learning obstacles of general users, resulting in low user participation and poor learning effect. It avoids the situation that users do not have enough exposure to content they are interested in and lack targeted guidance for difficult-to-master movements, and provides technical support for the popularization and dissemination of Peking Opera culture to non-professional groups, while improving the personalization and effectiveness of Peking Opera interactive experience.

[0145] Example 2, as shown in the appendix Figure 2 As shown, based on the inventive concept of a motion-capture-based interactive experience method for Peking Opera provided in Embodiment 1, this application also provides a motion-capture-based interactive experience device for Peking Opera, specifically including:

[0146] User action data acquisition module 01 is used to acquire user action data of multiple general experience users by combining motion capture, wherein the user action data includes limb action sequences and behavioral action sets.

[0147] The group dual-parameter calculation module 02 is used to calculate the group attention parameter and group learning difficulty parameter of multiple target Peking Opera stylized movements based on the user action data.

[0148] The high-value action slice extraction module 03 is used to extract high-value action slices from multiple target Peking Opera stylized actions based on the group attention parameter and the group learning difficulty parameter, and obtain an action slice sequence.

[0149] The experience content adjustment module 04 is used to dynamically adjust the experience content of subsequent Peking Opera interactive experiences based on the action slice sequence.

[0150] In one embodiment, the user action data acquisition module 01 is further configured to:

[0151] Based on historical Peking Opera interactive experience logs, the first action data and first operation data of users within the table are obtained; based on big data, the second action data and second operation data of users outside the table are obtained; motion capture is performed on the first action data and the second action data to obtain the body movement sequence of general users; using the target Peking Opera stylized action as an index, statistical analysis is performed on the first operation data and the second operation data to obtain the behavioral action set.

[0152] Furthermore, the user action data acquisition module 01 also includes:

[0153] The set of behavioral actions includes at least the number of clicks, single playback duration, and number of loop playbacks corresponding to multiple target Peking Opera stylized action videos.

[0154] In one embodiment, the population two-parameter calculation module 02 is further used for:

[0155] The group attention parameter is determined based on the interaction concentration index of general users towards the target Peking Opera stylized movements, including: statistically analyzing and obtaining the cumulative number of times general users selected the target Peking Opera stylized movements based on the body movement sequence; calculating the average single dwell time of general users in the imitation and practice stage of the target Peking Opera stylized movements based on the body movement sequence; statistically analyzing and obtaining the percentage of clicks by general users on the target Peking Opera stylized movements based on the set of behavioral actions; statistically analyzing and obtaining the typical action replay ratio of general users on the target Peking Opera stylized movements based on the set of behavioral actions; and performing a weighted calculation based on the cumulative number of times, the average single dwell time, the percentage of clicks, and the typical action replay ratio to obtain the interaction concentration index.

[0156] Furthermore, the population two-parameter calculation module 02 also includes:

[0157] The group learning difficulty parameter is determined based on the convergence of general experience users imitating the target Peking Opera stylized movements, including: extracting the deviation vector between each general experience user and the standard movement template when imitating the target Peking Opera stylized movements based on the body movement sequence; performing cluster analysis on the deviation vectors of multiple general experience users and calculating the silhouette coefficient of the clustering results; statistically obtaining the variance change trend of multiple deviation vectors over time; and calculating the movement convergence based on the silhouette coefficient and the variance change trend.

[0158] In one embodiment, the high-value action slice extraction module 03 is also used for:

[0159] The group attention parameter is compared with a first preset threshold, and the group learning difficulty parameter is compared with a second preset threshold. When the group attention parameter is higher than the first preset threshold, the corresponding target Peking Opera stylized movement is marked as a high-attractive segment. When the group learning difficulty parameter is higher than the second preset threshold, the corresponding target Peking Opera stylized movement is marked as a high-difficulty teaching segment. The first preset threshold and the second preset threshold are confidence thresholds based on statistical analysis.

[0160] In one embodiment, the experience content adjustment module 04 is also used for:

[0161] Based on the group attention parameters, the highly attractive slices are serialized, and in the Peking Opera interactive experience scene, the highly attractive slices are displayed at the top according to the serialization results; combined with big data, at least one auxiliary teaching resource package is generated and associated with each of the high-difficulty teaching slices, wherein the auxiliary teaching resource package includes at least motion trajectory decomposition animation, typical error prompts, and key posture freeze-frame explanations; when the high-difficulty teaching slice is selected as experience content, the auxiliary teaching resource package is invoked.

[0162] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0163] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0164] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and variations fall within the scope of this application and its equivalents, this application intends to include such modifications and variations.

Claims

1. A motion capture based interactive Peking Opera experience method, characterized in that, include: By combining motion capture, user action data of multiple general users are obtained, wherein the user action data includes limb movement sequences and behavioral action sets; Based on the user action data, calculate the group attention parameter and group learning difficulty parameter for multiple target Peking Opera stylized actions; Based on the group attention parameter and the group learning difficulty parameter, high-value action slices are extracted from multiple target Peking Opera stylized actions to obtain action slice sequences. Based on the aforementioned action slice sequence, the experience content of subsequent Peking Opera interactive experiences is dynamically adjusted; The group attention parameter is determined based on the interaction concentration index of general users towards the target Peking Opera stylized actions, including: Based on the aforementioned body movement sequence, statistical analysis was performed to obtain the cumulative number of times that general users selected the target Peking Opera stylized movements; Based on the aforementioned body movement sequence, the average single dwell time of general experience users in the imitation and practice of the target Peking Opera stylized movements is calculated; Based on the set of behavioral actions, statistical analysis was conducted to obtain the percentage of click actions by general users on the target stylized Peking Opera actions. Based on the set of behavioral actions, statistical analysis was conducted to obtain the typical action replay ratio of general experience users on the target stylized Peking Opera actions. The interaction concentration index is obtained by weighting the cumulative number of times, the average single dwell time, the percentage of click actions, and the typical action replay ratio. The group learning difficulty parameter is determined based on the convergence of generalized users imitating the target stylized Peking Opera movements, including: Based on the body movement sequence, extract the deviation vector between each general experience user and the standard movement template when imitating the target Peking Opera stylized movement; Cluster analysis is performed on the deviation vectors of multiple general experience users, and the silhouette coefficient of the clustering results is calculated; The variance of multiple deviation vectors is statistically analyzed to obtain the variance variation trend over time. The convergence of the action is calculated by weighting the profile coefficient and the variance variation trend.

2. The method for interactive Peking Opera experience based on motion capture as described in claim 1, characterized in that, By combining motion capture, user action data from multiple general users is acquired. This user action data includes sequences of limb movements and sets of behavioral actions, including: Based on historical Peking Opera interactive experience logs, obtain the first action data and first operation data of the users in the table; Based on big data, acquire the second action data and second operation data of users experiencing the off-table experience; Motion capture is performed on the first motion data and the second motion data to obtain the body motion sequence of the general user. Using the target stylized Peking Opera movements as an index, statistical analysis is performed on the first operation data and the second operation data to obtain the set of behavioral movements.

3. The motion capture-based Peking opera interactive experience method of claim 1, wherein, Based on the group attention parameter and the group learning difficulty parameter, high-value action slices are extracted from multiple target Peking Opera stylized movements to obtain action slice sequences, including: The group attention parameter is compared with a first preset threshold, and the group learning difficulty parameter is compared with a second preset threshold. When the group attention parameter is higher than the first preset threshold, the corresponding target Peking Opera stylized action is marked as a high-attractive slice; When the group learning difficulty parameter is higher than the second preset threshold, the corresponding target Peking Opera stylized movement is marked as a high-difficulty teaching segment; Wherein, the first preset threshold and the second preset threshold are confidence thresholds based on statistical analysis.

4. The motion capture-based Peking opera interactive experience method of claim 3, wherein, Based on the aforementioned action slice sequence, the content of subsequent Peking Opera interactive experiences is dynamically adjusted, including: Based on the group attention parameters, the highly attractive slices are serialized, and the highly attractive slices are displayed at the top in the Peking Opera interactive experience scene according to the serialization results; Based on big data, at least one auxiliary teaching resource package is generated and associated for each of the high-difficulty teaching segments. The auxiliary teaching resource package includes at least motion trajectory decomposition animation, typical error prompts and key posture freeze-frame explanations. When the advanced teaching segment is selected as the experience content, the auxiliary teaching resource package is invoked.

5. The motion capture-based Peking opera interactive experience method of claim 1, wherein, The set of behavioral actions includes at least the number of clicks, single playback duration, and number of loop playbacks corresponding to multiple target Peking Opera stylized action videos.

6. A motion capture based interactive Peking Opera experience device, characterized in that, The device is used to execute the Peking Opera interactive experience method based on motion capture as described in any one of claims 1-5, the device comprising: The user action data acquisition module is used to acquire user action data from multiple general users by combining motion capture. The user action data includes a sequence of limb movements and a set of behavioral actions. The group dual-parameter calculation module is used to calculate the group attention parameter and group learning difficulty parameter of multiple target Peking Opera stylized movements based on the user action data. The high-value action slice extraction module is used to extract high-value action slices from multiple target Peking Opera stylized actions based on the group attention parameter and the group learning difficulty parameter, and obtain an action slice sequence. The experience content adjustment module is used to dynamically adjust the experience content of subsequent Peking Opera interactive experiences based on the action slice sequence.

Images