Action data comparison method and device, storage medium and electronic equipment

Abstract

The invention discloses an action data comparison method and device, a storage medium and electronic equipment. The method comprises the steps that first action data and second action data are obtained, the first action data are to-be-compared data obtained after a target user executes a target action, and the second action data are standard data of the target action; under the condition that the target action belongs to the first service scene, comparing the first action data with the second action data through a first algorithm matched with the first service scene to obtain a first comparison result; and under the condition that the target action belongs to the second service scene, comparing the first action data with the second action data through a second algorithm matched with the second service scene to obtain a second comparison result. The technical problem that the comparison flexibility of the action data is low is solved.

Description

Technical Field

[0001] The present application relates to the field of computers, and in particular to a method, device, storage medium and electronic device for comparing motion data. Background Art

[0002] In the application scenarios of motion data comparison, due to the diversity of business scenarios, the comparison standards for motion data in each scenario are also different. However, the current related technologies often use too single methods when performing motion data comparison, and fail to adjust the corresponding comparison standards according to different business scenarios. This lack of targeted comparison method makes motion data comparison inflexible when adapting to changing business needs. Therefore, there is a problem of low flexibility in motion data comparison.

[0003] To address the above-mentioned problems, no effective solution has been proposed yet. Summary of the invention

[0004] The embodiments of the present application provide a method, device, storage medium and electronic device for comparing action data, so as to at least solve the technical problem of low flexibility in comparing action data.

[0005] According to one aspect of an embodiment of the present application, a method for comparing action data is provided, including: obtaining first action data and second action data, wherein the first action data is data to be compared obtained after analyzing a target user performing a target action, and the second action data is standard data of the target action; when the target action belongs to a first business scenario, the first action data is compared with the second action data by using a first algorithm that matches the first business scenario to obtain a first comparison result; when the target action belongs to a second business scenario, the first action data is compared with the second action data by using a second algorithm that matches the second business scenario to obtain a second comparison result.

[0006] According to another aspect of an embodiment of the present application, an action data comparison device is also provided, including: an acquisition unit, used to acquire first action data and second action data, wherein the first action data is the data to be compared obtained after analyzing the target user performing the target action, and the second action data is the standard data of the target action; a first comparison unit, used to compare the first action data with the second action data by a first algorithm matched by the first business scenario when the target action belongs to a first business scenario, to obtain a first comparison result; a second comparison unit, used to compare the first action data with the second action data by a second algorithm matched by the second business scenario when the target action belongs to a second business scenario, to obtain a second comparison result.

[0007] As an optional solution, the above-mentioned first business scenario is a business scenario in which the number of lateral actions is greater than or equal to a first preset threshold value, and the above-mentioned first comparison unit includes: a first comparison module, which is used to combine the two-dimensional coordinate recognition algorithm and the three-dimensional coordinate recognition algorithm to compare the above-mentioned first action data with the above-mentioned second action data to obtain the above-mentioned first comparison result.

[0008] As an optional solution, the above-mentioned first comparison module includes: a first processing submodule, which is used to use the above-mentioned two-dimensional coordinate recognition algorithm to process the above-mentioned first action data and the above-mentioned second action data to obtain the first coordinate data of the above-mentioned first action data on the two-dimensional plane, and the second coordinate data of the above-mentioned second action data on the two-dimensional plane; a first comparison submodule, which is used to compare the above-mentioned first coordinate data with the above-mentioned second coordinate data to obtain a first result; a second processing submodule, which is used to use the above-mentioned three-dimensional coordinate recognition algorithm to process the above-mentioned first action data and the above-mentioned second action data to obtain the third coordinate data of the above-mentioned first action data on the three-dimensional plane, and the fourth coordinate data of the above-mentioned second action data on the three-dimensional plane; a second comparison submodule, which is used to compare the above-mentioned third coordinate data with the above-mentioned fourth coordinate data to obtain a second result; an allocation submodule, which is used to allocate a first weight to the above-mentioned first result based on the number of the above-mentioned lateral actions, and allocate a second weight to the above-mentioned second result; an integration submodule, which is used to integrate the above-mentioned first result and the above-mentioned second result using the above-mentioned first weight and the above-mentioned second weight to obtain the above-mentioned first comparison result.

[0009] As an optional solution, the above-mentioned device also includes: a display unit, which is used to compare the above-mentioned first action data with the above-mentioned second action data through the first algorithm matched by the above-mentioned first business scenario to obtain a first comparison result, and then superimpose and display the above-mentioned first comparison result by drawing an action point map of the above-mentioned target user performing the above-mentioned target action and a standard action point map corresponding to the standard data of the above-mentioned target action.

[0010] As an optional solution, the above-mentioned device also includes: a prompt unit, which is used to provide feedback prompts for insufficient areas where the difference between the above-mentioned action point map and the above-mentioned standard action point map is greater than or equal to a second preset threshold value during the process of superimposing and displaying the above-mentioned first comparison result by drawing an action point map of the above-mentioned target user performing the above-mentioned target action and a fitness action point map corresponding to the standard data of the above-mentioned target action, wherein the above-mentioned feedback prompts are used to prompt the deficiencies and areas for improvement in the actions corresponding to the above-mentioned insufficient areas.

[0011] As an optional solution, the above-mentioned acquisition unit includes: a collection module, which is used to collect the target screen when the above-mentioned target user performs the above-mentioned target action; an analysis module, which is used to analyze the above-mentioned target screen to obtain the skeleton point coordinates of the above-mentioned target user, wherein the above-mentioned first action data includes the above-mentioned skeleton point coordinates; an extraction module, which is used to extract the standard skeleton point coordinates corresponding to the above-mentioned target action from a pre-stored standard action library, wherein the above-mentioned second action data includes the above-mentioned standard skeleton point coordinates.

[0012] As an optional solution, the second business scenario is a dance imitation scenario, and the second comparison unit includes: a second comparison module, which is used to obtain the second comparison result by comparing the position and movement trajectory of the key point in the first action data and the second action data.

[0013] As an optional solution, the second business scenario is an entertainment imitation scenario, and the second comparison unit includes: a third comparison module, which is used to use the cosine similarity algorithm of the overall score to compare the similarity between the first action data and the second action data to obtain the second comparison result.

[0014] According to another aspect of the embodiments of the present application, a computer program product or a computer program is provided, the computer program product or the computer program including computer instructions, the computer instructions being stored in a computer-readable storage medium. A processor of an electronic device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the electronic device performs the above action data comparison method.

[0015] According to another aspect of an embodiment of the present application, there is also provided an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the action data comparison method through the computer program.

[0016] In an embodiment of the present application, first action data and second action data are obtained, wherein the first action data is the data to be compared obtained after analyzing the target user performing the target action, and the second action data is the standard data of the target action; when the target action belongs to the first business scenario, the first action data is compared with the second action data through the first algorithm matched with the first business scenario to obtain a first comparison result; when the target action belongs to the second business scenario, the first action data is compared with the second action data through the second algorithm matched with the second business scenario to obtain a second comparison result.

[0017] First, the embodiment of the present application clearly distinguishes between the first action data and the second action data. The first action data represents the analysis result of the user's actual execution of the action, while the second action data serves as a standard reference. This clear division of data lays the foundation for subsequent flexible comparison according to different scenarios.

[0018] Secondly, the core of the embodiment of the present application is to select a suitable comparison algorithm according to the business scenario to which the target action belongs. For the first business scenario, a specially designed first algorithm is used for comparison, which fully considers the specific needs and comparison criteria under the scenario. Similarly, for the second business scenario, a second algorithm that matches it is used. The choice of this algorithm is not arbitrary, but is based on the results of in-depth understanding and analysis of the comparison requirements of each business scenario, thereby achieving the purpose of finding the most suitable comparison algorithm in different business scenarios, thereby obtaining more accurate and meaningful comparison results, thereby achieving the technical effect of improving the comparison flexibility of action data, and then solving the technical problem of low comparison flexibility of action data. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0020] Figure 1 is a schematic diagram of an application environment of an optional action data comparison method according to an embodiment of the present application;

[0021] Figure 2 is a schematic diagram of a process of an optional action data comparison method according to an embodiment of the present application;

[0022] Figure 3 is a schematic diagram of an optional action data comparison method according to an embodiment of the present application;

[0023] Figure 4 is a schematic diagram of another optional action data comparison method according to an embodiment of the present application;

[0024] Figure 5 is a schematic diagram of an optional motion data comparison device according to an embodiment of the present application;

[0025] Figure 6 It is a schematic diagram of the structure of an optional electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0026] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0027] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0028] According to one aspect of an embodiment of the present application, a method for comparing action data is provided. Optionally, as an optional implementation, the above-mentioned method for comparing action data can be applied to, but is not limited to, Figure 1 In the environment shown, the environment may include, but is not limited to, a user device 102 and a server 112 . The user device 102 may include, but is not limited to, a display 104 , a processor 106 , and a memory 108 . The server 112 includes a database 114 and a processing engine 116 .

[0029] The specific process can be as follows:

[0030] Step S102, the user device 102 acquires first action data and second action data, wherein the first action data is the data to be compared obtained after analyzing the target user performing the target action, and the second action data is the standard data of the target action;

[0031] Step S104, sending the first action data and the second action data to the server 112 via the network 110;

[0032] Steps S106-S108, the server 112 uses the processing engine 116 to compare the first action data with the second action data by using a first algorithm matching the first business scenario to obtain a first comparison result when the target action belongs to the first business scenario, or compares the first action data with the second action data by using a second algorithm matching the second business scenario to obtain a second comparison result when the target action belongs to the second business scenario;

[0033] Step S110 , sending the first comparison result or the second comparison result to the user device 102 via the network 110 , the user device 102 displays the first comparison result or the second comparison result on the display 104 via the processor 106 , and stores the first comparison result or the second comparison result in the memory 108 .

[0034] remove Figure 1 In addition to the examples shown, the terminal device may be a terminal device configured with a target client, which may include but is not limited to at least one of the following: a mobile phone, a laptop, a tablet computer, a PDA, a MID (Mobile Internet Devices), a PAD, a desktop computer, a smart TV, etc. The target client may be a video client, an instant messaging client, a browser client, an education client, etc. The network may include but is not limited to: a wired network (such as a local area network, a metropolitan area network, and a wide area network), a wireless network (such as Bluetooth, WIFI), and other networks that implement wireless communication. The server may be a single server, or a server cluster consisting of multiple servers, or a cloud server. The above is only an example, and this embodiment is not limited to this.

[0035] Optionally, as an optional implementation, as Figure 2 As shown, the motion data comparison method can be executed by an electronic device, which can be, for example, Figure 1 The user device or server shown in the figure comprises:

[0036] S202, obtaining first action data and second action data, wherein the first action data is data to be compared obtained after analyzing a target user performing a target action, and the second action data is standard data of the target action;

[0037] This step involves collecting data of the user's actions (first action data) and obtaining standard data of the actions (second action data). These data may include the speed, angle, trajectory, etc. of the action.

[0038] S204, when the target action belongs to the first business scenario, comparing the first action data with the second action data by using a first algorithm matching the first business scenario to obtain a first comparison result;

[0039] S206: When the target action belongs to the second business scenario, the first action data and the second action data are compared by using a second algorithm matching the second business scenario to obtain a second comparison result.

[0040] In an optional embodiment, the first action data refers to the data obtained by analyzing the target user performing a specific action (target action). Such data is usually collected and processed by sensors, cameras or other data capture devices for subsequent comparison and analysis. In short, it represents the data performance when the user actually performs the action.

[0041] In an optional embodiment, the second action data refers to standard or benchmark data of the target action, which is usually data generated by experts or certified standard actions. These data are used as a reference for comparison to evaluate the accuracy and quality of the user's action.

[0042] In an optional embodiment, the first business scenario and the second business scenario: these two terms refer to different application scenarios or contextual environments. In the context of action comparison, different business scenarios may represent different types of actions, different evaluation criteria, or different application purposes. For example, fitness training, dance teaching, sports competition, etc. may all be considered different business scenarios.

[0043] In an optional embodiment, the first algorithm and the second algorithm are comparison methods designed for different business scenarios. Each algorithm is optimized according to the requirements and characteristics of a specific scenario to ensure the accuracy and effectiveness of the comparison results.

[0044] In an optional embodiment, the first comparison result and the second comparison result: these refer to the conclusion or score obtained after comparing the first action data and the second action data through a specific algorithm. These results are usually used to evaluate the accuracy of the user's execution of the action and provide feedback or suggestions to improve the action.

[0045] In an optional embodiment, the business scenario to which the target action belongs is determined: based on the type of action performed by the user, determine to which business scenario it belongs (such as fitness, dance, etc.).

[0046] In an optional embodiment, a matching algorithm is selected for comparison: if the target action belongs to a first business scenario, a first algorithm matching the scenario is used for comparison. If the target action belongs to a second business scenario, a second algorithm matching the scenario is used for comparison.

[0047] In an optional embodiment, the first action data and the second action data are compared by a selected algorithm to obtain a comparison result, which can be a score, a series of indicators or specific improvement suggestions for guiding the user to improve his action execution.

[0048] It should be noted that in the process of motion data comparison, two core data sets are involved: first motion data and second motion data. First motion data refers to the data generated by the target user after performing a specific action obtained through technical means (such as smart device camera acquisition and AI analysis), which represents the user's actual action performance. In contrast, second motion data is the standard or ideal data for this specific action, usually used as a benchmark for comparison.

[0049] When it is necessary to compare these two sets of data, the solution takes into account that the actions may belong to different business scenarios, and these scenarios may have different requirements and evaluation criteria for the actions. Therefore, the solution proposes a strategy for selecting the corresponding comparison algorithm based on the business scenario. Specifically, if the target action belongs to the first business scenario (for example, a certain action in aerobics), the first algorithm that matches the scenario is used for comparison; if the target action belongs to the second business scenario (for example, a certain step in dance), the second algorithm is used. In this way, each business scenario can obtain a highly targeted and accurate comparison result.

[0050] To further illustrate, it is optional to assume that there is a fitness application through which users can follow fitness videos to exercise. The application captures the user's movements (first movement data) and compares them with the standard movements in the video (second movement data). If the user is doing the "mountain pose" in yoga, this movement requires high accuracy and stability of the posture and belongs to the first business scenario. At this time, the application will use the first algorithm, which may focus on comparing the differences between the user's movements and the standard movements in the positions of key skeletal points, as well as the stability of the posture.

[0051] If the user is learning the "locking" steps in street dance, this action focuses more on the sense of rhythm and the smoothness of the action, which belongs to the second business scenario. In this case, the application will switch to the second algorithm, which may focus more on comparing the synchronization of the user's action with the standard action in time series, as well as the naturalness of the action transition.

[0052] Through the embodiments provided by the present application, first action data and second action data are obtained, wherein the first action data is the data to be compared obtained after analyzing the target user performing the target action, and the second action data is the standard data of the target action; when the target action belongs to a first business scenario, the first action data and the second action data are compared by a first algorithm matching the first business scenario to obtain a first comparison result; when the target action belongs to a second business scenario, the first action data and the second action data are compared by a second algorithm matching the second business scenario to obtain a second comparison result.

[0053] First, the embodiment of the present application clearly distinguishes between the first action data and the second action data. The first action data represents the analysis result of the user's actual execution of the action, while the second action data serves as a standard reference. This clear division of data lays the foundation for subsequent flexible comparison according to different scenarios.

[0054] Secondly, the core of the embodiment of the present application is to select a suitable comparison algorithm according to the business scenario to which the target action belongs. For the first business scenario, a specially designed first algorithm is used for comparison, which fully considers the specific needs and comparison criteria in the scenario. Similarly, for the second business scenario, a second algorithm that matches it is used. The choice of this algorithm is not arbitrary, but is based on an in-depth understanding and analysis of the comparison requirements of each business scenario, thereby achieving the purpose of finding the most suitable comparison algorithm in different business scenarios, thereby obtaining a more accurate and meaningful comparison result, so as to achieve the technical effect of improving the flexibility of action data comparison.

[0055] As an optional solution, the first business scenario is a business scenario in which the number of side actions is greater than or equal to a first preset threshold value. The first action data is compared with the second action data through a first algorithm matched by the first business scenario to obtain a first comparison result, including: combining a two-dimensional coordinate recognition algorithm and a three-dimensional coordinate recognition algorithm to compare the first action data with the second action data to obtain a first comparison result.

[0056] In an optional embodiment, the first business scenario: The first business scenario specifically refers to business scenarios in which the number of side actions is greater than or equal to a preset threshold. Side actions can be understood as actions performed by a user when observed from the side. This scenario may appear in a variety of applications, such as fitness action displays from the side, dance steps from the side, etc. The first preset threshold: This is a pre-set value used to determine whether the number of side actions reaches or exceeds a certain standard, thereby determining whether the processing method of the first business scenario is applicable.

[0057] In an optional embodiment, the first algorithm: This is a comparison algorithm that matches the first business scenario. In this scenario, the first algorithm combines a two-dimensional coordinate recognition algorithm and a three-dimensional coordinate recognition algorithm to achieve a comparison between the first motion data and the second motion data. Two-dimensional coordinate recognition algorithm: This algorithm is generally used to process and analyze motion data on a two-dimensional plane, such as image data captured by a camera. It can identify key points in the image and convert them into two-dimensional coordinates for subsequent motion comparison. Three-dimensional coordinate recognition algorithm: Unlike a two-dimensional algorithm, a three-dimensional coordinate recognition algorithm is able to process and analyze motion data in three-dimensional space. This usually involves the use of a depth camera, sensor, or other device that can capture three-dimensional information. Through three-dimensional coordinates, the user's actual movements can be more accurately simulated and compared.

[0058] In an optional embodiment, the first comparison result: This is the result obtained after the first action data and the second action data are compared by the first algorithm. This result may be a score, a series of indicators or other forms of feedback for evaluating the accuracy and quality of the user's action.

[0059] To further illustrate, optionally assume that there is a smart fitness coaching application where a user can record a video of themselves performing side push-ups through the mobile phone camera. The application analyzes the movements in the video (first movement data) and compares it with the standard side push-up movements (second movement data). In this example, the first business scenario is a scenario where the number of side push-ups reaches or exceeds a preset threshold (such as 10). The application uses a first algorithm that combines two-dimensional and three-dimensional coordinate recognition algorithms to compare the user's movements. The two-dimensional algorithm can analyze key points in the video image, such as the position of the elbows, hips, and feet; while the three-dimensional algorithm can use depth information to more accurately evaluate the depth, speed, and stability of the user's movements. After the comparison is completed, the application will give a first comparison result, which may be a score (such as 90 points) and some specific improvement suggestions (such as "keep the elbow closer to the body").

[0060] By combining the recognition algorithms of two-dimensional and three-dimensional coordinates, the first algorithm can provide more comprehensive and accurate action comparison results. This not only helps users better understand their action performance, but also provides them with more targeted improvement suggestions. In addition, this flexible comparison mechanism is also applicable to a variety of different side action scenarios, thus enhancing the versatility and practicality of the application.

[0061] As an optional solution, combining the two-dimensional coordinate recognition algorithm and the three-dimensional coordinate recognition algorithm, comparing the first action data with the second action data, and obtaining a first comparison result, including:

[0062] S1-1, using a two-dimensional coordinate recognition algorithm to process the first motion data and the second motion data to obtain first coordinate data of the first motion data on the two-dimensional plane and second coordinate data of the second motion data on the two-dimensional plane;

[0063] S1-2, comparing the first coordinate data with the second coordinate data to obtain a first result;

[0064] S1-3, using a three-dimensional coordinate recognition algorithm to process the first motion data and the second motion data to obtain third coordinate data of the first motion data on the three-dimensional plane and fourth coordinate data of the second motion data on the three-dimensional plane;

[0065] S1-4, comparing the third coordinate data with the fourth coordinate data to obtain a second result;

[0066] S1-5, assigning a first weight to the first result and a second weight to the second result based on the number of side actions;

[0067] S1-6, using the first weight and the second weight, integrating the first result and the second result to obtain a first comparison result.

[0068] In an optional embodiment, the recognition algorithm processing of two-dimensional coordinates refers to processing the motion data using two-dimensional coordinate recognition technology and converting it into coordinate data on a two-dimensional plane. This can quantify the position and change of the motion in two-dimensional space. The recognition algorithm processing of three-dimensional coordinates is similar to two-dimensional, but the processed data contains depth information, which can more comprehensively reflect the state of the motion in three-dimensional space.

[0069] In an optional embodiment, the first result and the second result: are preliminary comparison results obtained by comparing the two-dimensional coordinate data and the three-dimensional coordinate data, respectively. They may be expressed in the form of similarity scores, difference metrics, or other forms. The first weight and the second weight: are weights assigned to the first result and the second result based on the number of side actions. The allocation of weights may take into account the importance of different dimensional data in a specific action. Integration: The first result and the second result are integrated into a comprehensive comparison result (first comparison result) using weights. This process may involve weighted averaging, weighted summation, or other mathematical operations.

[0070] It should be noted that this embodiment describes a detailed comparison process, by combining the two-dimensional coordinate recognition algorithm and the three-dimensional coordinate recognition algorithm, the first action data and the second action data are processed and compared, and finally a comprehensive comparison result (i.e., the first comparison result) is obtained. This process includes data processing, coordinate comparison, weight allocation and result integration.

[0071] To further illustrate, optionally, taking a dance teaching application as an example, the user records his or her dance moves (first movement data) through a camera, and the application compares it with the standard dance moves (second movement data). The application first uses a two-dimensional coordinate recognition algorithm to process the two dance moves, extracts the coordinate data of each on the two-dimensional plane (such as the position changes of the hands and feet), and compares to obtain a first result, such as the accuracy score of the hand and foot movement. Next, the application uses a three-dimensional coordinate recognition algorithm to process the same data, taking into account factors such as the depth and rotation of the dance moves, and compares to obtain a second result, such as the smoothness score of the body rotation. Considering that there are more side movements in this dance, the application may assign a higher weight (second weight) to the second result because it more comprehensively reflects the user's dance performance. Finally, the application combines the first result and the second result (using their respective weights) to obtain a comprehensive first comparison result for evaluating the overall performance of the user's dance.

[0072] Through the embodiments provided by this application, the advantages of two-dimensional and three-dimensional coordinate recognition algorithms are combined to more comprehensively evaluate the user's action performance. By assigning weights to the comparison results of different dimensions, the focus of the evaluation can be flexibly adjusted to adapt to different scenarios and needs. This comprehensive comparison method improves the accuracy and objectivity of action evaluation, helps users better understand their action performance, and make improvements based on feedback.

[0073] As an optional solution, after comparing the first action data with the second action data through the first algorithm for matching the first business scenario to obtain a first comparison result, the method further includes:

[0074] The first comparison result is displayed by overlaying an action point map of the target user performing the target action and a standard action point map corresponding to the standard data of the target action.

[0075] In an optional embodiment, an action point map is a visualization tool used to display the position, trajectory or state of key points during the execution of an action. In the fields of action analysis, motion capture, etc., point maps help to intuitively understand the structure and dynamic characteristics of actions. Overlay display refers to the display of two or more images, charts or data layers superimposed on each other. The purpose of overlay display is to compare the difference between the target user's action and the standard action, so as to more intuitively reflect the first comparison result.

[0076] It should be noted that this embodiment describes further operations taken after comparing two sets of action data (first action data and second action data) through a specific algorithm (i.e., a first algorithm for matching a first business scenario) and obtaining a first comparison result. Specifically, this method also includes drawing two action point maps - one generated by the target user performing the target action, and the other corresponding to the standard data of the target action - and superimposing and displaying the two point maps.

[0077] To further illustrate, it is optional to take sports training as an example, assuming that the first business scenario is tennis serving action analysis. The coach wants to evaluate whether the student's serving action is standard. First, the coach uses a special algorithm (first algorithm) to compare the student's serving action data (first action data) with the standard serving action data (second action data) to obtain a preliminary comparison result (first comparison result). Next, in order to more intuitively display the comparison results, the coach draws a point map of the student's serving action (showing key information such as hitting point and body posture) and a point map of the standard serving action. Finally, the coach superimposes the two point maps together for display. By comparison, you can clearly see which parts of the student's serving process are consistent with the standard action and which parts have deviations.

[0078] By drawing and superimposing action point maps, this method not only provides quantitative comparison results, but also adds intuitive visualization effects. This is extremely useful for coaches, athletes, or any scenario that requires analysis of action accuracy. Superimposed display can quickly reveal the advantages and disadvantages of action execution, thereby guiding users to make targeted improvements. In addition, this visualization method can also be used as an auxiliary tool in the teaching or training process to help improve learning efficiency and training effects.

[0079] As an optional scheme, in the process of superimposing and displaying the first comparison result by drawing an action point map of the target user performing the target action and a fitness action point map corresponding to the standard data of the target action, the method also includes: providing feedback prompts for insufficient areas where the difference between the action point map and the standard action point map is greater than or equal to a second preset threshold, wherein the feedback prompts are used to prompt the deficiencies and areas for improvement in the actions corresponding to the insufficient areas.

[0080] It should be noted that this embodiment further refines the process of superimposing and displaying the action point map mentioned above. When superimposing and displaying the action point map of the target user and the point map of the standard action, the method also includes a feedback mechanism for identifying and prompting the difference between the two. Specifically, when the difference between the action point map and the standard action point map is greater than or equal to a preset threshold (the second preset threshold), the system will provide feedback prompts for these areas with significant differences (i.e., insufficient areas) to clearly point out the deficiencies in the action and areas that need improvement.

[0081] To further illustrate, optionally taking the squat action in fitness training as an example, when the user performs squat exercises, his movements are captured and converted into action point maps. At the same time, the system also has a point map of the standard squat action. When the user's squat action point map is superimposed and displayed with the standard squat action point map, the system begins to analyze the differences between the two. If the system detects that the user's knee bending angle, back inclination or foot position are significantly different from the standard action, and these differences exceed the preset second preset threshold, the system will give feedback prompts in these areas where the differences are obvious. These feedback prompts may appear in the form of highlights, color markings or text descriptions, aiming to inform users of their specific deficiencies in the squat action, such as "excessive knee extension", "excessive back inclination", etc., and prompt them how to make adjustments to improve the action.

[0082] By introducing a difference detection and feedback prompt mechanism, this method not only provides a visual comparison of actions, but also implements targeted action guidance. This instant feedback can help users more accurately identify and correct their shortcomings in the execution of actions, thereby effectively improving training results and reducing potential risks caused by incorrect actions. At the same time, this personalized feedback method also enhances the interactivity between users and the training system and improves the user experience.

[0083] As an optional solution, obtaining the first action data and the second action data includes:

[0084] S2-1, collecting the target screen when the target user performs the target action;

[0085] S2-2, analyzing the target image to obtain the skeleton point coordinates of the target user, wherein the first action data includes the skeleton point coordinates;

[0086] S2-3, extracting standard skeleton point coordinates corresponding to the target action from a pre-stored standard action library, wherein the second action data includes the standard skeleton point coordinates.

[0087] In an optional embodiment, target screen: refers to the video or image captured and recorded when the target user performs the target action. Skeletal point coordinates: by analyzing the target screen, the key skeletal points (such as elbows, knees, neck, etc.) of the target user's body can be identified and located, and the position information (i.e., coordinates) of these points constitute the skeletal point coordinates. Standard action library: a database that pre-stores various standard action data, which contains standard skeletal point coordinates of various actions.

[0088] It should be noted that this embodiment describes the process of obtaining the first action data and the second action data. The first action data is derived from the screen analysis when the target user performs the target action, and the user's skeletal point coordinates are obtained by analyzing these screens. The second action data is extracted from a pre-stored standard action library, and the standard skeletal point coordinates corresponding to the target action.

[0089] To further illustrate, we can optionally take yoga practice as an example, assuming that the target user is doing a "cat-cow pose". First, a camera or other sensor is used to capture the image of the user doing this action (i.e., the target image). Then, image recognition technology is used to analyze the target image, identify the user's various skeletal points (such as head, shoulders, elbows, etc.) and determine their coordinate positions, which constitute the first action data. Next, the standard skeletal point coordinates of the "cat-cow pose" are retrieved from the standard action library, and these coordinates are used as the second action data. The first action data and the second action data obtained in this way provide accurate basic data for subsequent action comparison and analysis. By comparing the user's actual skeletal point coordinates with the standard skeletal point coordinates, the accuracy and standardization of the user's actions can be accurately evaluated, and targeted feedback and guidance can be provided to the user.

[0090] In addition, this method is not only suitable for yoga practice, but can also be widely used in sports training, dance teaching, rehabilitation therapy and other fields to help users improve their skills, prevent sports injuries and promote health recovery.

[0091] As an optional solution, the second business scenario is a dance imitation scenario. The first action data and the second action data are compared through a second algorithm matched by the second business scenario to obtain a second comparison result, including: obtaining the second comparison result by comparing the position and movement trajectory of key points in the first action data and the second action data.

[0092] In an optional embodiment, a dance imitation scenario refers to a scenario in which a user attempts to imitate a standard dance move or dance sequence. A second algorithm is an algorithm designed specifically for a dance imitation scenario, which is used to compare the dance moves imitated by the user (first movement data) with standard dance moves (second movement data). Key points: In dance movements, key points usually refer to important parts of the body or key nodes of movement, such as wrists, ankles, waist, etc. Their positions and movement trajectories are crucial to defining and evaluating dance movements. A second comparison result is the result obtained by comparing the first movement data and the second movement data through the second algorithm, which reflects the accuracy and similarity of the user's imitation of the dance.

[0093] It should be noted that this embodiment describes how to compare the first motion data with the second motion data in a dance imitation scenario (second business scenario) and obtain a second comparison result. Specifically, it involves comparing the positions and movement trajectories of key points in the two sets of motion data through a specific algorithm (second algorithm).

[0094] To further illustrate, optionally assume that the user is imitating a classic dance move called "Moon Walk". First, the system captures the screen of the user performing this move, and obtains the key point position and movement trajectory of the user's skeleton through analysis, which is the first action data. Next, the system extracts the standard key point position and movement trajectory of "Moon Walk" from the standard action library as the second action data. Then, a second algorithm is used to compare the key point positions and movement trajectories in the two sets of data. For example, the algorithm may check whether the user's footsteps match the footsteps in the standard action, or whether the user's arm swings are consistent with the arm swings in the standard action. Finally, the system will give a second comparison result based on the comparison result, such as a score or a set of feedback, telling the user the accuracy of their imitation and where they need to improve.

[0095] Through the embodiments provided in this application, this key point-based comparison method is not only applicable to a single dance move, but can also be extended to the comparison of the entire dance sequence. By continuously capturing and analyzing the user's dance moves and comparing them with the standard dance sequence, the system can provide the user with more comprehensive feedback and guidance.

[0096] In addition, the effectiveness of this method lies in its objectivity and accuracy. By quantitatively comparing the positions and movement trajectories of key points, the system can accurately evaluate the similarity and accuracy of the user's imitation of the dance, thereby helping the user to better learn and improve dance skills. At the same time, this comparison method can also provide dance teachers or judges with an objective and unified evaluation standard, making dance teaching and competitions more fair and efficient.

[0097] As an optional solution, the second business scenario is an entertainment imitation scenario. The first action data and the second action data are compared through a second algorithm matched with the second business scenario to obtain a second comparison result, including: using a cosine similarity algorithm of an overall score to compare the similarity between the first action data and the second action data to obtain a second comparison result.

[0098] In an optional embodiment, an entertainment imitation scenario refers to a scenario in which a user imitates a specific action or behavior for the purpose of entertainment. Cosine similarity algorithm for overall scoring: This is a mathematical algorithm used to quantify the similarity between two vectors (action data in this context). It evaluates their similarity by calculating the cosine value of the angle between the two vectors. The closer the cosine value is to 1, the more similar the two vectors (i.e., action data) are.

[0099] It should be noted that in the description of this embodiment, the "second business scenario" is specifically referred to as the "entertainment imitation scenario". In this scenario, there is a specific comparison process, that is, a certain "second algorithm" is used to compare the "first action data" and the "second action data" to generate a "second comparison result". The core of this comparison process is to use the "cosine similarity algorithm of overall score" to evaluate the similarity between the two sets of action data.

[0100] To further illustrate, let's assume that there is an online dance teaching platform where users can watch professional dance videos and try to imitate them. In this entertainment imitation scenario, the user's imitation dance video (second action data) will be captured and compared with the original professional dance video (first action data). The platform uses the cosine similarity algorithm to analyze the action data in the two videos and give an overall score that reflects the accuracy of the user's imitation.

[0101] By using the cosine similarity algorithm of the overall score, the similarity between the first action data and the second action data can be accurately compared in the entertainment imitation scene. This comparison method is not only objective and accurate, but also can provide users with instant feedback, so that they can clearly know the effect of their imitation. This instant feedback mechanism helps to enhance the user's sense of participation and achievement, thereby enhancing the attractiveness and interactivity of the entire entertainment imitation activity.

[0102] As an optional solution, for ease of understanding, the above motion data comparison method is applied to AI-assisted fitness scenarios. The core of AI-assisted fitness lies in the limb recognition function that relies on artificial intelligence technology. At present, limb recognition technology based on artificial intelligence has become increasingly mature and easy to implement. Through smart devices equipped with cameras, combined with advanced artificial intelligence frameworks such as ML kit, OpenPose, AlphaPose, etc., the user's limb bone point data can be captured and analyzed in real time to provide accurate guidance for fitness exercises.

[0103] Typical AI-assisted fitness applications usually use artificial intelligence technology to identify the user's movements and obtain their skeletal point data. These applications then compare the user's point data with the point data of standard movements to evaluate the similarity of the movements. At present, there are various algorithms for comparing movement similarity, including cosine similarity comparison, key angle comparison, deep learning methods based on big data, and OKS (Object Keypoint Similarity). These methods have their own characteristics and there is no absolute superiority or inferiority. The choice of scoring method should be determined based on the processing power of the limb recognition device and the business purpose. For example, if the business goal is to provide users with accurate movement suggestions, an estimation algorithm that focuses on the similarity of local limb details should be used; if the business only needs to motivate users to participate in exercise without overly precise movement feedback, then the overall data can be used for evaluation without in-depth calculation of the similarity details of local limbs. This flexible selection mechanism ensures that AI-assisted fitness applications can meet the needs of different users and business scenarios.

[0104] In the development process of AI-assisted fitness applications, the coordinate data of human skeleton points obtained through artificial intelligence limb recognition technology is not intuitive for developers. The simple coordinate value cannot directly map the specific action form, and it is even more difficult to directly compare the difference between two actions through coordinate data. Therefore, developers need to use advanced visualization tools or algorithms to convert these coordinate data into a more understandable graphical representation in order to accurately evaluate and analyze the execution of actions.

[0105] There are many methods for evaluating action similarity, including fine algorithms based on local point details, macroscopic evaluation of overall point similarity, and analysis methods focusing on limb angles and action vectors. These methods each have their own focus and advantages, and the choice of which method depends on the specific application requirements and scenarios. At the same time, differences in acquisition equipment will also affect the accuracy of the identified action point data. Different devices may have different recognition accuracy and stability. Therefore, when applying different algorithms, corresponding adjustments and optimizations need to be made according to the characteristics of the device to ensure the accuracy and reliability of the evaluation results.

[0106] In the development and testing process of AI-assisted fitness applications, the subjective differences of testers are an issue that cannot be ignored. Due to the lack of unified and objective evaluation criteria, different testers may give different similarity evaluations for the same action. In order to solve this problem, this embodiment needs to establish a set of scientific, objective, and quantifiable evaluation index systems, combining a variety of evaluation methods and data analysis techniques to conduct a comprehensive and accurate assessment of action similarity. This can not only reduce the impact of subjective differences on the evaluation results, but also provide strong data support for the continuous optimization and iteration of the application.

[0107] In view of the above technical problems, this embodiment proposes the following technical solutions:

[0108] Using professional drawing tools, this embodiment draws the user's actual action data and the standard fitness action data into images, and superimposes the two images for display. Through this visualization method, this embodiment can intuitively observe the difference between the two actions, including key elements such as posture, angle and amplitude, thereby providing a clear and objective comparison basis for developers and testers.

[0109] This embodiment will display the results calculated by various similarity algorithms to ensure the comprehensiveness and transparency of the data. For similarity algorithms that focus on local limb details, this embodiment will list the scores of each part in detail and mark each local limb with a clear number on the original image to facilitate quick positioning and searching. Such a design not only helps this embodiment to fully understand the similarities and differences between actions, but also helps this embodiment to select the most suitable algorithm for application according to the needs of the current business scenario.

[0110] Through the above technical solution, this embodiment can display the recognized user action images and the calculation results of various algorithms in an intuitive and objective way. This provides a reliable data reference for testers, enabling them to give relatively accurate test result feedback based on this objective information. The entire solution not only improves the transparency and efficiency of data processing, but also enhances the practicality and user experience of AI-assisted fitness applications.

[0111] It should be noted that during the development phase of the AI-assisted fitness function, this embodiment uses an innovative method for action comparison and algorithm selection. First, the user's actions following the fitness video are captured by a smartphone, and then the skeletal point information of the user's actions is accurately extracted using advanced artificial intelligence tools. In order to ensure the accuracy of the comparison, this embodiment accurately aligns the user's point information with the timeline of the fitness action, and performs appropriate scaling and normalization.

[0112] Next, this embodiment draws the point map of the user action and the point map of the standard fitness action respectively, and further overlays and displays the two point maps. This overlay comparison method makes the difference between the user action and the standard action clear at a glance, providing developers with an intuitive and efficient comparison tool.

[0113] More importantly, this embodiment simultaneously displays the results of multiple similarity evaluation algorithms below the comparison chart. This includes not only the comparison between different algorithms, but also the comparison of the effects of the same algorithm under different parameter settings. This design enables developers to comprehensively and objectively evaluate the applicability of various algorithms under current business goals.

[0114] In general, this embodiment provides a comprehensive and objective decision support tool for developers of AI-assisted fitness-related applications by intuitively displaying the differences between user actions and standard actions, as well as comparing the results of multiple similarity algorithms. This not only helps them choose the similarity algorithm that best suits their current business goals, but also guides them to fine-tune the algorithm parameters, thereby improving the accuracy and user experience of AI-assisted fitness functions.

[0115] To further illustrate, the optional Figure 3 As shown, the specific steps are as follows:

[0116] S302, data collection: the user follows the fitness video to exercise, and the smart device camera collects and recognizes the user's movements in real time, and then saves the collected data as a log for subsequent processing;

[0117] Specifically, the user follows the fitness video to exercise, and the camera of the smart device is used to collect the user's action picture in real time. The user's action is analyzed using artificial intelligence tools including but not limited to ML kit to obtain the user's skeletal point coordinates, and the skeletal point coordinates are saved. The recognition result is the three-dimensional coordinate value of each key point, where the x and y coordinates are the pixel positions of the input image, and the z-axis is the estimated value of the AI ​​tool based on the learning data, which is not as accurate as the x and y axis data. The accuracy is related to the quality and content of the provided recognition picture. The accuracy of the z-axis coordinate and whether it can be applied to the current business scenario can also be compared and evaluated in step 5.

[0118] S304, data processing: processing the collected log data, including time alignment of the current user action data to be compared with the standard action data, and coordinate normalization to ensure data consistency and comparability;

[0119] Specifically, the saved user point coordinate data is scaled so that the pixel area of ​​the recognized human body is consistent with the pixel area of ​​the human body in the fitness video, and then the coordinates are normalized, and the coordinate representation range is normalized to the range of 0 to 1 using the L2 norm. Finally, the user action and the action extracted from the fitness video are time-aligned using the DTW (Dynamic Time Warping) algorithm to obtain the fitness action corresponding to each user action data, and a user action data and the corresponding fitness action are regarded as a set of comparison data.

[0120] S306, data drawing: based on the processed data, draw a user action point map and a standard action point map, and overlay and compare the two images to intuitively show the difference between the user action and the standard action;

[0121] Specifically, a drawing tool including but not limited to Python's PIL library is used to draw the user action point data into a point map (such as Figure 4 The corresponding fitness movements and the point data of the fitness movements are plotted into a point map (as shown in the figure on the left). Figure 4 Finally, the user point map and the corresponding fitness action point map are superimposed and displayed simultaneously (as shown in the middle figure). Figure 4 As shown in the figure on the right, the difference between the two actions can be compared intuitively.

[0122] S308, data evaluation: quantitatively evaluate the user's actions by calculating multiple action similarity evaluation indicators, such as position deviation, speed matching, etc., and display the evaluation results below the comparison chart to facilitate the user to intuitively understand the quality of their own actions;

[0123] Use multiple evaluation indicators to calculate the similarity between user actions and fitness actions, and use some more conventional action scoring algorithms. For example, calculate the limb angle composed of several key bone nodes to calculate the 2D plane angle and 3D space angle, so that the evaluation accuracy of the two angles can be compared. Although deep learning tools such as ML kit can identify the three-dimensional coordinates of the bone points, the accuracy of the z-axis coordinates is not as good as the x- and y-axis coordinates. It is not convenient to identify the angle when the limbs overlap only with the x- and y-axis coordinates. Both have their own advantages and disadvantages; In addition to the scoring algorithm based on angles, there are also scoring algorithms based on key points. The common one is the cosine similarity algorithm, which treats the identified bone point data as a high-dimensional vector and directly compares the cosine similarity of the standard action point and the user action point. This method is simple and easy to use, but it is directly calculated using all the point data, and it is difficult to obtain the local similarity of the limbs; the similarity estimation method based on key points also has the OKS (Object KeypointSimilarity) method, which calculates the similarity of each normalized key point coordinate. This method can be specific to each point to compare the similarity, but does not consider the relative position relationship between points. In addition to a variety of similarity comparison algorithms, these algorithms often have many adjustable parameters. For example, the OKS algorithm based on limb angles adjusts the weight based on the difficulty of labeling each point, and the similarity calculation algorithm based on limb angles can adjust the score coefficient of the limb angle difference to adjust the overall score. The evaluation method is mainly based on whether the scores of superimposed image differences or overall similarity comparisons, or point position comparisons, or angle difference comparisons are consistent with general cognition. For example, when the z-axis recognition coordinates are inaccurate, the corresponding 3D space angle calculation is inaccurate, and the score is also inaccurate. The actual action difference and the scored action difference can be intuitively seen through the comparison chart.

[0124] S310, comparison decision: Based on the evaluation results, determine whether it is necessary to continue to compare other action data. If necessary, use the next action data for comparison; if not, proceed to the next step;

[0125] S312, algorithm selection: based on the comparison results of all action data, select the optimal scoring algorithm or algorithm parameters to ensure the accuracy and effectiveness of subsequent action evaluation;

[0126] Specifically, S304 to S310 are repeated to screen similarity evaluation algorithms using multiple sets of user comparison data, and a similarity estimation algorithm that most testers consider reasonable is selected through communication with the testers, and the process is terminated. Specifically, the criteria for screening algorithms vary according to different business scenarios. If a standard fitness action video contains a lot of side content, more points will be blocked if only 2D coordinates are used for identification. Using 3D coordinates will have a better effect, but 3D coordinates themselves are not very accurate, so a weighted balance between the two is required. The weighted coefficient adjustment effect needs to be actually listed and referenced together with the picture for effective comparison. If it is a fast-movement dance video imitation scene, the point-based similarity calculation algorithm may be more suitable than the angle-based calculation algorithm, because users may find it difficult to grasp the body angle information during fast movement. For fitness videos, higher requirements are placed on the user's body movements, and evaluation algorithms based on body angles are more suitable. For entertainment-related imitation movements, which do not require too high accuracy, it is more appropriate to use the cosine similarity algorithm for overall scoring. In addition to the selection of algorithms with different characteristics required for different business scenarios, the selection of different parameters for the same algorithm can also be compared through this tool. For example, the evaluation algorithm based on body angles determines the score based on the size of the angle difference, and the angle difference tolerance range can be adjusted to modify the looseness of the score.

[0127] Through the embodiments provided in this application, advanced technical means are used to accurately draw user point maps and standard action point maps, and then overlay and compare the two, so as to intuitively reveal the subtle differences between user actions and standard actions. This method not only provides an intuitive basis for judging the effects of various similarity evaluation algorithms, but also helps developers to have a deeper understanding of the accuracy of action recognition and the direction of improvement.

[0128] In addition, the embodiment of the present application provides a comprehensive display of evaluation indicators to ensure that developers can select the most appropriate evaluation algorithm according to the specific needs of the current business scenario. By displaying each evaluation indicator in detail, a transparent and objective decision support platform is provided for developers, which helps them make wise choices among many algorithms.

[0129] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present application.

[0130] According to another aspect of the embodiment of the present application, a motion data comparison device for implementing the above-mentioned motion data comparison method is also provided. Figure 5 As shown, the device comprises:

[0131] An acquisition unit 502 is used to acquire first action data and second action data, wherein the first action data is data to be compared obtained after analyzing a target user performing a target action, and the second action data is standard data of the target action;

[0132] A first comparison unit 504 is used to compare the first action data with the second action data by using a first algorithm for matching the first business scenario to obtain a first comparison result when the target action belongs to the first business scenario;

[0133] The second comparison unit 506 is used to compare the first action data with the second action data by using a second algorithm matching the second business scenario to obtain a second comparison result when the target action belongs to the second business scenario.

[0134] For specific embodiments, reference may be made to the examples shown in the above-mentioned method for comparing action data, which will not be described in detail in this example.

[0135] As an optional solution, the first business scenario is a business scenario in which the number of side actions is greater than or equal to a first preset threshold, and the first comparison unit 504 includes:

[0136] The first comparison module is used to combine the two-dimensional coordinate recognition algorithm and the three-dimensional coordinate recognition algorithm to compare the first action data with the second action data to obtain a first comparison result.

[0137] For specific embodiments, reference may be made to the examples shown in the above-mentioned method for comparing action data, which will not be described in detail in this example.

[0138] As an optional solution, the first comparison module includes:

[0139] A first processing submodule, used to process the first motion data and the second motion data by using a two-dimensional coordinate recognition algorithm to obtain first coordinate data of the first motion data on the two-dimensional plane and second coordinate data of the second motion data on the two-dimensional plane;

[0140] A first comparison submodule, used for comparing the first coordinate data with the second coordinate data to obtain a first result;

[0141] A second processing submodule is used to process the first motion data and the second motion data by using a three-dimensional coordinate recognition algorithm to obtain third coordinate data of the first motion data on the three-dimensional plane and fourth coordinate data of the second motion data on the three-dimensional plane;

[0142] A second comparison submodule, used for comparing the third coordinate data with the fourth coordinate data to obtain a second result;

[0143] an allocating submodule for allocating a first weight to the first result and a second weight to the second result based on the number of side actions;

[0144] The integration submodule is used to integrate the first result and the second result by using the first weight and the second weight to obtain a first comparison result.

[0145] For specific embodiments, reference may be made to the examples shown in the above-mentioned method for comparing action data, which will not be described in detail in this example.

[0146] As an optional solution, the device also includes: a display unit, which is used to compare the first action data with the second action data through a first algorithm matched with a first business scenario to obtain a first comparison result, and then superimpose and display the first comparison result by drawing an action point map of the target user performing the target action and a standard action point map corresponding to the standard data of the target action.

[0147] For specific embodiments, reference may be made to the examples shown in the above-mentioned method for comparing action data, which will not be described in detail in this example.

[0148] As an optional solution, the device also includes: a prompt unit, which is used to provide feedback prompts for insufficient areas where the difference between the action point map and the standard action point map is greater than or equal to a second preset threshold value during the process of superimposing and displaying the first comparison result by drawing an action point map of the target user performing the target action and a fitness action point map corresponding to the standard data of the target action, wherein the feedback prompts are used to prompt the deficiencies and areas for improvement in the actions corresponding to the insufficient areas.

[0149] For specific embodiments, reference may be made to the examples shown in the above-mentioned method for comparing action data, which will not be described in detail in this example.

[0150] As an optional solution, the acquisition unit 502 includes:

[0151] A collection module, used to collect the target screen when the target user performs the target action;

[0152] An analysis module, used for analyzing the target picture to obtain the skeleton point coordinates of the target user, wherein the first action data includes the skeleton point coordinates;

[0153] The extraction module is used to extract standard skeleton point coordinates corresponding to the target action from a pre-stored standard action library, wherein the second action data includes the standard skeleton point coordinates.

[0154] For specific embodiments, reference may be made to the examples shown in the above-mentioned method for comparing action data, which will not be described in detail in this example.

[0155] As an optional solution, the second business scenario is a dance imitation scenario, and the second comparison unit 506 includes: a second comparison module, which is used to obtain a second comparison result by comparing the position and movement trajectory of the key point in the first action data and the second action data.

[0156] For specific embodiments, reference may be made to the examples shown in the above-mentioned method for comparing action data, which will not be described in detail in this example.

[0157] As an optional solution, the second business scenario is an entertainment imitation scenario, and the second comparison unit 506 includes: a third comparison module, which is used to use the cosine similarity algorithm of the overall score to compare the similarity between the first action data and the second action data to obtain a second comparison result.

[0158] For specific embodiments, reference may be made to the examples shown in the above-mentioned method for comparing action data, which will not be described in detail in this example.

[0159] According to another aspect of the embodiment of the present application, an electronic device for implementing the above-mentioned action data comparison method is also provided. The electronic device can be but is not limited to: Figure 1 The user device 102 or the server 112 shown in FIG. 1 is used as an example to illustrate the electronic device as the user device 102. Figure 6 As shown, the electronic device includes a memory 602 and a processor 604. The memory 602 stores a computer program, and the processor 604 is configured to execute the steps in any of the above method embodiments through the computer program.

[0160] In an optional embodiment, the electronic device may be located in at least one network device among a plurality of network devices of a computer network.

[0161] In an optional embodiment, the processor may be configured to perform the following steps through a computer program:

[0162] S1, obtaining first action data and second action data, wherein the first action data is data to be compared obtained after analyzing a target user performing a target action, and the second action data is standard data of the target action;

[0163] S2, when the target action belongs to the first business scenario, comparing the first action data with the second action data by using a first algorithm matching the first business scenario to obtain a first comparison result;

[0164] S3, when the target action belongs to the second business scenario, the first action data and the second action data are compared by a second algorithm matching the second business scenario to obtain a second comparison result.

[0165] Alternatively, a person skilled in the art may understand that: Figure 6 The structure shown is for illustration only. Figure 6 The structure of the electronic device is not limited. Figure 6 More or fewer components (such as network interfaces, etc.) as shown in, or with Figure 6 Different configurations are shown.

[0166] Among them, the memory 602 can be used to store software programs and modules, such as the program instructions / modules corresponding to the action data comparison method and device in the embodiment of the present application. The processor 604 executes various functional applications and data processing by running the software programs and modules stored in the memory 602, that is, realizing the above-mentioned action data comparison method. The memory 602 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 602 may further include a memory remotely arranged relative to the processor 604, and these remote memories may be connected to the electronic device via a network. Among them, the memory 602 may be specifically, but not limited to, used to store information such as the first action data and the second action data. As an example, if Figure 6 As shown, the memory 602 may include but is not limited to the first acquisition unit 502, the first comparison unit 504 and the second comparison unit 506 in the action data comparison device. In addition, it may also include but is not limited to other module units in the action data comparison device, which will not be repeated in this example.

[0167] Optionally, the transmission device 606 is used to receive or send data via a network. In one example, the transmission device 606 includes a network adapter (Network Interface Controller, NIC), which can be connected to other network devices and routers via a network cable so as to communicate with the Internet or a local area network. In one example, the transmission device 606 is a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0168] In addition, the electronic device further includes: a display 608 for displaying information such as the first action data and the second action data; and a connection bus 610 for connecting various module components in the electronic device.

[0169] In other embodiments, the user device or server may be a node in a distributed system, wherein the distributed system may be a blockchain system, and the blockchain system may be a distributed system formed by connecting the multiple nodes through network communication. Among them, the nodes may form a peer-to-peer network, and any form of computing device, such as a server, user device, or other electronic device, may become a node in the blockchain system by joining the peer-to-peer network.

[0170] According to one aspect of the present application, a computer program product is provided, the computer program product comprising a computer program / instruction, the computer program / instruction comprising a program code for executing the method shown in the flow chart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication part, and / or installed from a removable medium. When the computer program is executed by a central processing unit, various functions provided by the embodiments of the present application are executed.

[0171] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0172] It should be noted that the computer system of the electronic device is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0173] The computer system includes a central processing unit (CPU), which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) or the program loaded from the storage part into the random access memory (RAM). In the random access memory, various programs and data required for the system operation are also stored. The central processing unit, the read-only memory and the random access memory are connected to each other through a bus. The input / output interface (Input / Output interface, i.e., I / O interface) is also connected to the bus. The following components are connected to the input / output interface: an input part including a keyboard, a mouse, etc.; an output part including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage part including a hard disk, etc.; and a communication part including a network interface card such as a local area network card, a modem, etc. The communication part performs communication processing via a network such as the Internet. The drive is also connected to the input / output interface as needed. Removable media, such as disks, optical disks, magneto-optical disks, semiconductor memories, etc., are installed on the drive as needed so that the computer program read therefrom is installed into the storage part as needed.

[0174] In particular, according to an embodiment of the present application, the process described in each method flow chart can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a computer readable medium, and the computer program contains a program code for executing the method shown in the flow chart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication part, and / or installed from a removable medium. When the computer program is executed by a central processing unit, various functions defined in the system of the present application are executed.

[0175] According to one aspect of the present application, a computer-readable storage medium is provided, and a processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the methods provided in the above-mentioned various optional implementations.

[0176] In an optional embodiment, the computer-readable storage medium may be configured to store a computer program for performing the following steps:

[0177] S1, obtaining first action data and second action data, wherein the first action data is data to be compared obtained after analyzing a target user performing a target action, and the second action data is standard data of the target action;

[0178] S2, when the target action belongs to the first business scenario, comparing the first action data with the second action data by using a first algorithm matching the first business scenario to obtain a first comparison result;

[0179] S3, when the target action belongs to the second business scenario, the first action data and the second action data are compared by a second algorithm matching the second business scenario to obtain a second comparison result.

[0180] Optionally, in the embodiments of the present application, the term "module" or "unit" refers to a computer program or a part of a computer program with a predetermined function, and works together with other related parts to achieve a predetermined goal, and can be implemented in whole or in part by using software, hardware (such as processing circuits or memories), or a combination thereof. Similarly, a processor (or multiple processors or memories) can be used to implement one or more modules or units. In addition, each module or unit can be part of an overall module or unit that includes the function of the module or unit.

[0181] In an optional embodiment, a person of ordinary skill in the art may understand that all or part of the steps in the various methods of the above-described embodiments may be completed by instructing hardware related to the electronic device through a program, and the program may be stored in a computer-readable storage medium, and the storage medium may include: a flash drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0182] If the integrated unit in the above embodiment is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in the above computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for one or more computer devices to execute all or part of the steps of the methods of each embodiment of the present application.

[0183] In the above embodiments of the present application, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0184] In addition, the device embodiments described above are only illustrative, such as the division of units, which is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. And the mutual coupling or direct coupling or communication connection shown or discussed may be through some interfaces, indirect coupling or communication connection of units or modules, which may be electrical or other forms.

[0185] The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, each functional unit in each embodiment of the present application may be integrated into a processing unit, or each unit may exist physically separately, or multiple units may be integrated into one unit. The above-mentioned integrated units may be implemented in the form of hardware or software functional units.

[0186] The above is only a preferred implementation of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A method for comparing motion data, characterized in that: include: Acquire first action data and second action data, wherein the first action data is data to be compared obtained after analyzing a target user performing a target action, and the second action data is standard data of the target action; When the target action belongs to a first business scenario, comparing the first action data with the second action data by using a first algorithm matching the first business scenario to obtain a first comparison result; In a case where the target action belongs to a second business scenario, the first action data and the second action data are compared by using a second algorithm matched with the second business scenario to obtain a second comparison result.

2. The method according to claim 1, characterized in that: The first business scenario is a business scenario in which the number of side actions is greater than or equal to a first preset threshold, and the first algorithm for matching the first business scenario compares the first action data with the second action data to obtain a first comparison result, including: The first action data and the second action data are compared by combining a two-dimensional coordinate recognition algorithm and a three-dimensional coordinate recognition algorithm to obtain the first comparison result.

3. The method according to claim 2, characterized in that The combining of the two-dimensional coordinate recognition algorithm and the three-dimensional coordinate recognition algorithm to compare the first action data with the second action data to obtain the first comparison result includes: Processing the first motion data and the second motion data by using the two-dimensional coordinate recognition algorithm to obtain first coordinate data of the first motion data on the two-dimensional plane and second coordinate data of the second motion data on the two-dimensional plane; Comparing the first coordinate data with the second coordinate data to obtain a first result; Processing the first motion data and the second motion data by using the three-dimensional coordinate recognition algorithm to obtain third coordinate data of the first motion data on the three-dimensional plane and fourth coordinate data of the second motion data on the three-dimensional plane; Comparing the third coordinate data with the fourth coordinate data to obtain a second result; assigning a first weight to the first result and a second weight to the second result based on the number of the side actions; The first result and the second result are integrated using the first weight and the second weight to obtain the first comparison result.

4. The method according to claim 1, characterized in that: After the first action data and the second action data are compared by the first algorithm for matching the first business scenario to obtain a first comparison result, the method further includes: The first comparison result is displayed by superimposing an action point map of the target user performing the target action and a standard action point map corresponding to the standard data of the target action.

5. The method according to claim 4, characterized in that In the process of superimposing and displaying the first comparison result by drawing an action point map of the target user performing the target action and a fitness action point map corresponding to the standard data of the target action, the method further includes: Feedback is provided for the deficient area where the difference between the action point map and the standard action point map is greater than or equal to a second preset threshold, wherein the feedback is used to indicate deficiencies and areas for improvement in the action corresponding to the deficient area.

6. The method according to any one of claims 1 to 5, characterized in that The obtaining of the first action data and the second action data comprises: Capturing a target screen when the target user performs the target action; Analyze the target screen to obtain the skeletal point coordinates of the target user, wherein the first action data includes the skeletal point coordinates; The standard skeleton point coordinates corresponding to the target action are extracted from a pre-stored standard action library, wherein the second action data includes the standard skeleton point coordinates.

7. The method according to any one of claims 1 to 5, characterized in that The second business scenario is a dance imitation scenario, and the second algorithm matched by the second business scenario compares the first action data with the second action data to obtain a second comparison result, including: The second comparison result is obtained by comparing the position and movement trajectory of the key point in the first action data and the second action data.

8. The method according to any one of claims 1 to 5, characterized in that The second business scenario is an entertainment imitation scenario, and the second algorithm matched by the second business scenario compares the first action data with the second action data to obtain a second comparison result, including: The similarity between the first action data and the second action data is compared using a cosine similarity algorithm of an overall score to obtain the second comparison result.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein the program executes the method according to any one of claims 1 to 8 when executed by an electronic device.

10. An electronic device comprising a memory and a processor, characterized in that: A computer program is stored in the memory, and the processor is configured to execute the method according to any one of claims 1 to 8 through the computer program.

Images