Dance movement scoring method, device, equipment and storage medium

Through human posture recognition and comparison of bone angle changes, accurate scoring of dance movements is achieved, the problem of inaccurate movement scores in the existing technology is solved, and a convenient learning assistance tool and objective judgment reference is provided.

CN114333046BActive Publication Date: 2024-11-26TSINGHUA UNIVERSITY
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202111358056.0
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2021-11-16
Publication Date
2024-11-26
Estimated Expiration
2041-11-16

AI Technical Summary

Technical Problem

The existing technology cannot accurately compare and score dance movements, making it difficult for dance students and coaches to accurately judge whether the movements are in place.

Method used

The three-dimensional coordinates of the dance movement are obtained through the human posture recognition model, the change in the angle of the bone is calculated, and compared with the standard movements. Differentiated scores and display pages are generated according to the scoring rules to achieve accurate comparison and scoring of the movements.

Benefits of technology

It provides a more accurate dance scoring system to help dance learners correct their movements and provide objective scoring methods, which are suitable for the evaluation reference of dance exam ratings.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114333046B_ABST
    Figure CN114333046B_ABST
Patent Text Reader

Abstract

The present invention provides a dance action scoring method, device, equipment and storage medium, including: inputting the human posture recognition data of the current student's dance action into a human body recognition model to obtain the first three-dimensional coordinates of the human body joints; obtaining the first angle between two connected bones of the human body in the same frame image according to the first three-dimensional coordinates; obtaining the first angle change of the first angle between two adjacent frames of images; comparing the first angle change with the standard angle change of the standard action to obtain the angle change difference, wherein the first angle change matches the action sequence fragment of the dance action where the corresponding standard angle change is located; obtaining the differentiated scores of different parts of the human body of the current student according to the angle change difference and the scoring rules corresponding to the current dance type. The technical solution of the present invention can score dance actions more accurately.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a dance movement scoring method, device, electronic equipment and a non-transient computer-readable storage medium. Background Art

[0002] As a fitness sport, dance has gradually entered the daily life of more and more people. During the current dance training process, dance students need to use a mirror to observe whether their dance movements are in place, so as to make self-adjustments, or the dance coach can judge whether the dance students' movements are in place.

[0003] However, neither the dance students themselves nor the dance coaches can accurately compare the dance moves of the dance students with the standard dance moves. Summary of the invention

[0004] The present invention provides a dance movement scoring method, device, electronic device and non-transitory computer-readable storage medium, which are used to solve the problem that dance movements cannot be accurately compared in the current technology and achieve accurate scoring of dance movements.

[0005] The present invention provides a dance action scoring method, comprising: inputting human posture recognition data of a current student's dance action into a human body recognition model to obtain first three-dimensional coordinates of human body joints; obtaining a first angle between two connected bones of a human body in the same frame image according to the first three-dimensional coordinates; obtaining a first angle change of the first angle between two adjacent frames of images; comparing the first angle change with a standard angle change of a standard action to obtain an angle change difference, wherein the first angle change matches an action sequence segment of a dance action where the corresponding standard angle change is located; and obtaining differentiated scores of different parts of the human body of the current student according to the angle change difference and a scoring rule corresponding to the current dance type.

[0006] According to a dance movement scoring method provided by the present invention, after obtaining differentiated scores of different parts of the body of the current student according to the angle change difference and the scoring rules corresponding to the current dance type, the method also includes: generating a front-end display page according to the angle change difference and the standard movement data of the standard movement, so as to compare and display the dance movement of the current student with the standard movement.

[0007] According to a dance movement scoring method provided by the present invention, before inputting the human posture recognition data of the current student's dance movements into a human posture recognition model, the method also includes: inputting the student dance video of the current student into a preset human posture recognition model to obtain the human posture recognition data.

[0008] According to a dance movement scoring method provided by the present invention, the front-end display page is generated according to the angle change difference and the standard movement data of the standard movement, including: normalizing the angle change difference to obtain a normalized difference; generating a line-dot diagram according to the normalized difference and the standard movement data using a drawing protocol, and the dots of the line-dot diagram are superimposed on corresponding positions of corresponding frames of a standard video of the standard movement.

[0009] According to a dance movement scoring method provided by the present invention, the method of generating a line dot diagram using a drawing protocol based on the normalized difference and the standard movement data includes: changing the display colors of the dots and lines of the line dot diagram based on the angle change difference and the set color rule.

[0010] According to a dance movement scoring method provided by the present invention, before inputting the student dance video of the current student into a preset human posture recognition model, the method further includes: obtaining the student dance video obtained by video capturing the dance movements of the current student.

[0011] According to a dance movement scoring method provided by the present invention, before comparing the first angle change with the standard angle change of the standard movement, the method also includes: obtaining a standard dance video obtained by video capture of the dance movements of a standard dancer; inputting human posture recognition data of the dance movements of the standard dancer obtained from the standard dance video into a human body recognition model to obtain second three-dimensional coordinates of human joints; obtaining a second angle between two connected bones of a human body in the same frame image according to the second three-dimensional coordinates; and obtaining the standard angle change of the second angle between two adjacent frames of images.

[0012] The present invention provides a dance action scoring device, comprising: a first recognition unit, used for inputting human body posture recognition data of a current student's dance action into a human body recognition model to obtain first three-dimensional coordinates of a human body joint; a first acquisition unit, used for acquiring a first angle between two connected bones of a human body in the same frame image according to the first three-dimensional coordinates; a second acquisition unit, used for acquiring a first angle change of the first angle between two adjacent frames of images; a comparison unit, used for comparing the first angle change with a standard angle change of a standard action to obtain an angle change difference, wherein the first angle change matches an action sequence segment of a dance action where the corresponding standard angle change is located; and a scoring unit, used for acquiring differentiated scores of different parts of the human body of the current student according to the angle change difference and a scoring rule corresponding to the current dance type.

[0013] According to a dance movement scoring device provided by the present invention, the device also includes a display unit, which is used to generate a front-end display page according to the angle change difference and the standard movement data of the standard movement, so as to compare and display the dance movement of the current student with the standard movement.

[0014] According to a dance movement scoring device provided by the present invention, the device also includes a second recognition unit, which is used to input the student dance video of the current student into a preset human posture recognition model to obtain the human posture recognition data.

[0015] According to a dance movement scoring device provided by the present invention, the display unit is also used to normalize the angle change difference to obtain a normalized difference; generate a line-dot diagram using a drawing protocol based on the normalized difference and the standard movement data, and the dots of the line-dot diagram are superimposed on corresponding positions of corresponding frames of a standard video of the standard movement.

[0016] According to a dance movement scoring device provided by the present invention, the display unit is further used to change the display colors of the dots and lines of the line-dot diagram according to the angle variation difference and the set color rule.

[0017] According to a dance movement scoring device provided by the present invention, the device also includes a third acquisition unit, which is used to acquire the student's dance video obtained by capturing the dance movements of the current student.

[0018] According to a dance movement scoring device provided by the present invention, the device also includes a standard unit, which is used to obtain a standard dance video obtained by video capturing the dance movements of a standard dancer; input human posture recognition data of the dance movements of the standard dancer obtained from the standard dance video into a human body recognition model to obtain second three-dimensional coordinates of human joints; obtain a second angle between two connected bones of a human body in the same frame image according to the second three-dimensional coordinates; and obtain a standard angle change of the second angle between two adjacent frames of images.

[0019] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of any of the above-mentioned dance movement scoring methods are implemented.

[0020] The present invention also provides a non-transitory computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of any of the above-mentioned dance movement scoring methods are implemented.

[0021] The dance action scoring method, device, electronic device and non-transitory computer-readable storage medium provided by the present invention can obtain the position coordinates of the human joints by performing human posture recognition and human body recognition on the current student's dance action, and can obtain the bone angle change in two adjacent frames of images according to the position coordinates of different human joints, and can compare the bone angle change with the standard angle change, and can obtain the current dance action score in combination with the dance scoring rules. This scheme can obtain a relatively accurate dance score.

[0022] The technical solution of the embodiment of the present invention provides a convenient and intelligent auxiliary tool for dance learning. The computer automatically compares and scores the movements based on the video information or the captured images. In most cases where there is no on-site guidance from a dance coach, the dance learning of students can be made more convenient and the error correction effect can be better. In addition, the technical solution of the embodiment of the present invention provides an objective auxiliary scoring method for dance examinations and grading. For some basic dance movements, an objective scoring mechanism is provided on the basis of manual scoring, which adds a reference method for judging for the organizers of dance examinations and grading. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0024] Figure 1 It is a flow chart of the dance movement scoring method provided by the present invention;

[0025] Figure 2 It is a schematic diagram of the process of obtaining standard action data provided by the present invention;

[0026] Figure 3 It is a structural schematic diagram of a dance movement scoring device provided by the present invention;

[0027] Figure 4 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0028] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0029] The terms used in one or more embodiments of the present invention are only for the purpose of describing specific embodiments, and are not intended to limit one or more embodiments of the present invention. The singular forms of "a", "said" and "the" used in one or more embodiments of the present invention and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used in one or more embodiments of the present invention refers to and includes any or all possible combinations of one or more associated listed items.

[0030] It should be understood that, although the terms first, second, etc. may be used to describe various information in one or more embodiments of the present invention, these information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of one or more embodiments of the present invention, the first may also be referred to as the second, and similarly, the second may also be referred to as the first. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0031] Exemplary embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0032] like Figure 1 FIG. 2 is a flow chart of a method for scoring dance movements according to an embodiment of the present invention. The method provided by the embodiment of the present invention can be executed by any electronic device with computer processing capabilities, such as a terminal or a server. Figure 1 As shown, the dance action scoring method includes:

[0033] Step 102, inputting the human body posture recognition data of the current student's dance movements into a human body recognition model to obtain the first three-dimensional coordinates of the human body joints.

[0034] Specifically, human posture recognition data refers to recognition data obtained by performing human posture recognition on a human body in an image or video through a human posture recognition model. Human posture recognition may be human key point posture recognition, which detects the position of key parts of human posture in each frame during the expression of human behavior, simplifies human posture into human key points, and classifies and recognizes the semantics of human posture expression through these key points. The key points in the embodiment of the present invention may be human joints. The human posture recognition model may be a neural network model. The human recognition model is a three-dimensional human body model, such as an SMPL (A Skinned Multi-Person Linear Model) model. The SMPL model is a vertex-based model with skin and texture, which can be easily adjusted to adapt to various human body shapes and postures.

[0035] Step 104: Acquire a first angle between two connected bones of a human body in the same frame image according to the first three-dimensional coordinates.

[0036] Specifically, the bones are connected through human joints, and the straight line where the bones are located can be obtained through the coordinates of the human joints, and then the angle between the straight lines where the two bones are located can be obtained.

[0037] Step 106: Obtain a first angle change of a first angle between two adjacent frames of images.

[0038] Specifically, the first angle variation can be obtained according to the first angle between two bones in a frame of image and the first angle between the two bones in a frame of image next to the frame of image.

[0039] Step 108, comparing the first angle variation with the standard angle variation of the standard movement to obtain an angle variation difference, wherein the first angle variation matches the action sequence segment of the dance movement where the corresponding standard angle variation is located.

[0040] Specifically, the standard angle variation is obtained based on a video of the same dance movements of a standard dancer, and the steps of obtaining the angle variation difference from the video are similar to steps 102 to 108 .

[0041] Step 110, obtaining differentiated scores for different parts of the body of the current student according to the angle variation difference and the scoring rule corresponding to the current dance type.

[0042] Specifically, different dance types allow different differences in the angle changes. According to the scoring rules corresponding to the current dance type, the score of the current student's dance movements can be obtained, that is, the above-mentioned differentiated score.

[0043] In the related art, the comparison between dance movements and standard movements is mostly done manually, and the amplitude and angle of the dance movements are determined manually by visual inspection, which is not very accurate, and the comparison standard is relatively subjective, so that the score obtained by the comparison is not accurate enough. In the technical solution of the embodiment of the present invention, the dance movements are analyzed through computer vision technology and motion analysis technology, and the difference in the angle change obtained by the analysis is scored by comparing it with the standard movement, so that a more accurate quantitative score and error correction guidance result can be obtained.

[0044] Before step 102, a student dance video obtained by video capturing the dance movements of the current student can be obtained, and the student dance video of the current student is input into a preset human posture recognition model to obtain human posture recognition data.

[0045] Specifically, the video acquisition device can be a camera or a wearable device. A color depth camera can be used to capture human dance movements, or a portable wearable device with the function of capturing color depth images can be used to capture human dance movements. The lens of the video acquisition device is controlled by a computer, and the dancer stands in the field of view of the video acquisition device. The video acquisition device can capture the dancer's dance movements to form a video stream. The video stream is input into the human posture recognition model, and the two-dimensional and three-dimensional parameters of the human body movement can be obtained. After that, the captured two-dimensional and three-dimensional parameters can be predicted and optimized using the human motion sequence coding engine.

[0046] In step 102, the three-dimensional coordinates corresponding to the human body joints can be calculated based on the optimized two-dimensional and three-dimensional parameters of the human body movement.

[0047] In step 104, the angle between every two bones of the human body in each frame of image can be calculated based on the three-dimensional coordinates of the human body joints, and then in step 106, the change of the angle between two frames is calculated.

[0048] Before step 108, it is also necessary to obtain the standard angle variation. The process includes the following steps: Figure 2 The following steps are shown:

[0049] Step 202, obtaining a standard dance video obtained by capturing dance movements of a standard dancer.

[0050] Step 204: input the human body posture recognition data of the standard dancer's dance movements obtained from the standard dance video into the human body recognition model to obtain the second three-dimensional coordinates of the human body joints.

[0051] Step 206, obtaining a second angle between two connected bones of the human body in the same frame image according to the second three-dimensional coordinates.

[0052] Step 208: Obtain the standard angle variation of the second angle between two adjacent frames of images.

[0053] In addition, when comparing the difference of the angle variation in step 108, it is required to align the time of the first angle variation with the standard angle variation. In an embodiment of the present invention, the Rule-Base method can be used to segment the human dance action fragments. Thus, the action sequences of the dance student and the dance coach are semantically segmented, the action sequence fragments are stripped out and the same action sequence fragments of the dance student and the dance coach are matched, that is, the time alignment of the first angle variation with the corresponding standard angle variation is achieved.

[0054] In the angle comparison link in step 108, the angle changes of the dance student and the dance coach can be compared frame by frame for the stripped action clips to obtain the angle change difference.

[0055] In the action comparison display link in step 110, a front-end display page can be generated according to the angle change difference and the standard action data of the standard action to compare and display the current student's dance action with the standard action.

[0056] Specifically, the angle variation difference can be normalized to obtain a normalized difference; a line dot diagram is generated based on the normalized difference and the standard action data using a drawing protocol, and the dots of the line dot diagram are superimposed on the corresponding positions of the corresponding frames of the standard video of the standard action. The dots in the line dot diagram represent nodes, and the lines represent bones.

[0057] During normalization, the angle change difference is mapped to a value between -1 and 1, recorded as diff(k), where k is the current frame number. diff(k) is sent from the server to the front-end program, which uses OpenGL or WebGL to render nodes and bones on the screen with dots and lines, and superimposes them on the corresponding positions of the human body in the video to achieve action comparison display. Among them, OpenGL is a cross-language, cross-platform application programming interface for rendering 2D (two-dimensional) and 3D (three-dimensional) vector graphics. WebGL is a drawing protocol. The line dot diagram can be generated using either OpenGL or WebGL.

[0058] In addition, the display colors of the dots and lines of the line dot diagram can be changed according to the angle variation difference and the set color rule. Specifically, the normalized difference diff(k) can be multiplied by the value corresponding to the RGB color value to change the display colors of the nodes and bones to provide a more intuitive display image.

[0059] The dance movement scoring scheme of the embodiment of the present invention realizes the semantic segmentation of the dance movement itself and the time calibration alignment through the human posture recognition link, the human recognition model establishment link, the three-dimensional coordinate calculation link, the joint angle and dynamic information calculation link, the cross-time matching and analysis link of the motion sequence, the dynamic information storage and call link, the movement comparison and scoring link and the client display rendering link, so as to obtain more accurate scoring data.

[0060] In an embodiment of the present invention, a dance learner and a dance coach respectively use a system and a product equipped with the technology of an embodiment of the present invention to record their own dance moves. The algorithm will match and find differences based on the movements of the two people, and provide quantitative guidance results for scoring, evaluation, and error correction, thereby helping students learn dance moves.

[0061] The technical solution of the embodiment of the present invention provides a convenient and intelligent auxiliary tool for dance learning. The computer automatically compares and scores the movements based on the video information or the captured images. In most cases where there is no on-site guidance from a dance coach, the dance learning of students can be made more convenient and the error correction effect can be better. In addition, the technical solution of the embodiment of the present invention provides an objective auxiliary scoring method for dance examinations and grading. For some basic dance movements, an objective scoring mechanism is provided on the basis of manual scoring, which adds a reference method for judging for the organizers of dance examinations and grading.

[0062] The dance action scoring method provided by the present invention obtains the position coordinates of the human joints by performing human posture recognition and human body recognition on the current student's dance action, and obtains the bone angle variation in two adjacent frames of images according to the position coordinates of different human joints, compares the bone angle variation with the standard angle variation, and combines the dance scoring rules to obtain the current dance action score. This scheme can obtain a relatively accurate dance score.

[0063] The dance movement scoring device provided by the present invention is described below. The dance movement scoring device described below and the dance movement scoring method described above can be referred to each other.

[0064] like Figure 3 As shown, a dance action scoring device according to an embodiment of the present invention comprises:

[0065] The first recognition unit 302 can be used to input the human body posture recognition data of the current student's dance movements into the human body recognition model to obtain the first three-dimensional coordinates of the human body joints.

[0066] Specifically, the human body recognition model is a three-dimensional human body model, such as an SMPL (ASkinned Multi-Person Linear Model) model. The SMPL model is a vertex-based model with skin and texture, which can be easily adjusted to adapt to various human body shapes and postures.

[0067] The first acquisition unit 304 may be configured to acquire a first angle between two connected bones of a human body in the same frame image according to the first three-dimensional coordinates.

[0068] Specifically, the straight line where the bones are located can be obtained through the coordinates of the human body joints, and then the angle between the straight lines where the two bones are located can be obtained.

[0069] The second acquisition unit 306 may be configured to acquire a first angle change of a first angle between two adjacent frames of images.

[0070] Specifically, the first angle variation can be obtained according to the first angle between two bones in a frame of image and the first angle between the two bones in a frame of image next to the frame of image.

[0071] The comparison unit 308 can be used to compare the first angle variation with the standard angle variation of the standard action to obtain an angle variation difference, wherein the first angle variation matches the action sequence segment of the dance action where the corresponding standard angle variation is located.

[0072] The scoring unit 310 may be used to obtain differentiated scores for different parts of the body of the current student according to the angle variation difference and the scoring rule corresponding to the current dance type.

[0073] In the related art, the comparison between dance movements and standard movements is mostly done manually, and the amplitude and angle of the dance movements are determined manually by visual inspection, which is not very accurate, and the comparison standard is relatively subjective, so that the score obtained by the comparison is not accurate enough. In the technical solution of the embodiment of the present invention, the dance movements are analyzed through computer vision technology and motion analysis technology, and the difference in the angle change obtained by the analysis is scored by comparing it with the standard movement, so that a more accurate quantitative score and error correction guidance result can be obtained.

[0074] In an embodiment of the present invention, the device may further include a display unit for generating a front-end display page according to the angle variation difference and the standard movement data of the standard movement, so as to compare and display the dance movement of the current student with the standard movement.

[0075] In an embodiment of the present invention, the device may further include a second recognition unit, which is used to input the student dance video of the current student into a preset human posture recognition model to obtain human posture recognition data.

[0076] The key points in the embodiments of the invention may be human joints. The human posture recognition model may be a neural network model.

[0077] In an embodiment of the present invention, the display unit can also be used to normalize the angle change difference to obtain a normalized difference; generate a line-dot diagram using a drawing protocol based on the normalized difference and standard action data, and the dots of the line-dot diagram are superimposed on the corresponding positions of the corresponding frames of the standard video of the standard action.

[0078] During normalization, the angle change difference is mapped to a value between -1 and 1, recorded as diff(k), where k is the current frame number. diff(k) is sent from the server to the front-end program, which uses OpenGL or WebGL to render nodes and bones on the screen as dots and lines, and superimposes them on the corresponding positions of the human body in the video to achieve action comparison display.

[0079] In the embodiment of the present invention, the display unit may also be used to change the display colors of the dots and lines of the line-dot diagram according to the angle variation difference and the set color rule.

[0080] In the embodiment of the present invention, the device may further include a third acquisition unit, which is used to acquire a student's dance video obtained by capturing the dance movements of the current student.

[0081] The camera of the video acquisition device is controlled by a computer. When the dancer stands in the field of view of the video acquisition device, the video acquisition device can capture the dancer's dance movements and form a video stream. The video stream is input into the human posture recognition model to obtain the two-dimensional and three-dimensional parameters of human motion. The human motion sequence coding engine is then used to predict and optimize the captured two-dimensional and three-dimensional parameters.

[0082] In an embodiment of the present invention, the device may also include a standard unit for obtaining a standard dance video obtained by capturing the dance movements of a standard dancer; inputting the human posture recognition data of the standard dancer's dance movements obtained from the standard dance video into a human body recognition model to obtain second three-dimensional coordinates of human joints; obtaining a second angle between two connected bones of a human body in the same frame image according to the second three-dimensional coordinates; and obtaining a standard angle change of the second angle between two adjacent frames of images.

[0083] In addition, when the comparison unit compares the difference in the angle variation, it is required to align the time of the first angle variation with the standard angle variation. In an embodiment of the present invention, the Rule-Base method can be used to segment the human dance action fragments. Thereby, the action sequences of the dance student and the dance coach are semantically segmented, the action sequence fragments are stripped out and the same action sequence fragments of the dance student and the dance coach are matched, that is, the time alignment of the first angle variation with the corresponding standard angle variation is achieved.

[0084] The technical solution of the embodiment of the present invention provides a convenient and intelligent auxiliary tool for dance learning. The computer automatically compares and scores the movements based on the video information or the captured images. In most cases where there is no on-site guidance from a dance coach, the dance learning of students can be made more convenient and the error correction effect can be better. In addition, the technical solution of the embodiment of the present invention provides an objective auxiliary scoring method for dance examinations and grading. For some basic dance movements, an objective scoring mechanism is provided on the basis of manual scoring, which adds a reference method for judging for the organizers of dance examinations and grading.

[0085] Since the functional modules of the dance movement scoring device of the exemplary embodiment of the present invention correspond to the steps of the exemplary embodiment of the dance movement scoring method described above, for details not disclosed in the device embodiment of the present invention, please refer to the embodiment of the dance movement scoring method described above.

[0086] The dance action scoring device provided by the present invention obtains the position coordinates of the human joints by performing human posture recognition and human body recognition on the current student's dance action. The bone angle change in two adjacent frames of images can be obtained according to the position coordinates of different human joints. The bone angle change is compared with the standard angle change, and the current dance action score can be obtained by combining the dance scoring rules. This solution can obtain a relatively accurate dance score.

[0087] Figure 4 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 4 As shown, the electronic device may include: a processor 410, a communication interface 420, a memory 430 and a communication bus 440, wherein the processor 410, the communication interface 420 and the memory 430 communicate with each other through the communication bus 440. The processor 410 may call the logic instructions in the memory 430 to execute the dance action scoring method, which includes: inputting the human posture recognition data of the dance action of the current student into the human body recognition model to obtain the first three-dimensional coordinates of the human body joints; obtaining the first angle between two connected bones of the human body in the same frame image according to the first three-dimensional coordinates; obtaining the first angle change between two adjacent frames of images; comparing the first angle change with the standard angle change of the standard action to obtain the angle change difference, wherein the first angle change matches the action sequence fragment of the dance action where the corresponding standard angle change is located; obtaining the differentiated scores of different parts of the human body of the current student according to the angle change difference and the scoring rule corresponding to the current dance type.

[0088] In addition, the logic instructions in the above-mentioned memory 430 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0089] On the other hand, the present invention also provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the dance movement scoring method provided by the above-mentioned methods, and the method includes: inputting the human posture recognition data of the current student's dance movements into a human recognition model to obtain a first three-dimensional coordinate of a human joint; obtaining a first angle between two connected bones of a human body in the same frame image according to the first three-dimensional coordinate; obtaining a first angle change of the first angle between two adjacent frames of images; comparing the first angle change with the standard angle change of a standard movement to obtain an angle change difference, wherein the first angle change matches the action sequence segment of the dance movement where the corresponding standard angle change is located; obtaining differentiated scores for different parts of the current student's body according to the angle change difference and the scoring rules corresponding to the current dance type.

[0090] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the above-mentioned dance movement scoring methods, the method comprising: inputting the human posture recognition data of the current student's dance movements into a human recognition model to obtain a first three-dimensional coordinate of a human joint; obtaining a first angle between two connected bones of a human body in the same frame image according to the first three-dimensional coordinate; obtaining a first angle change of the first angle between two adjacent frames of images; comparing the first angle change with a standard angle change of a standard movement to obtain an angle change difference, wherein the first angle change matches the action sequence segment of the dance movement in which the corresponding standard angle change is located; obtaining differentiated scores for different parts of the current student's body according to the angle change difference and the scoring rules corresponding to the current dance type.

[0091] The device embodiments described above are merely illustrative, wherein 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 modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0092] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0093] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A dance movement scoring method, characterized in that: include: Inputting the human body posture recognition data of the current student's dance movements into the human body recognition model to obtain the first three-dimensional coordinates of the human body joints; Acquire a first angle between two connected bones of a human body in the same frame image according to the first three-dimensional coordinates; Obtaining a first angle change of the first angle between two adjacent frames of images; Compare the first angle variation with the standard angle variation of the corresponding standard action to obtain an angle variation difference, wherein the first angle variation matches the action sequence segment of the dance action in which the corresponding standard angle variation is located; when comparing the angle variation difference, align the time of the first angle variation with the standard angle variation; Obtaining differentiated scores for different parts of the body of the current student according to the angle variation difference and the scoring rule corresponding to the current dance type; Generate a front-end display page according to the angle variation difference and the standard action data of the standard action to compare and display the dance action of the current student with the standard action; The step of generating a front-end display page according to the angle variation difference and the standard action data of the standard action includes: Normalizing the angle variation difference to obtain a normalized difference; Generate a line-dot diagram using a drawing protocol according to the normalized difference and the standard action data, wherein the dots of the line-dot diagram are superimposed on corresponding positions of corresponding frames of the standard video of the standard action; The display colors of the dots and lines of the line-dot diagram are changed according to the angle variation difference and the set color rule.

2. The method according to claim 1, characterized in that Before inputting the human body posture recognition data of the current student's dance movements into the human body recognition model, the method further comprises: The student dance video of the current student is input into a preset human posture recognition model to obtain the human posture recognition data.

3. The method according to claim 2, characterized in that Before inputting the student dance video of the current student into the preset human posture recognition model, the method further includes: Acquire the student's dance video obtained by capturing the dance movements of the current student.

4. The method according to claim 1, characterized in that: Before comparing the first angle variation with a standard angle variation of a corresponding standard action, the method further includes: Obtain a standard dance video obtained by capturing dance movements of a standard dancer; Inputting the human body posture recognition data of the dance movements of the standard dancer obtained from the standard dance video into the human body recognition model to obtain the second three-dimensional coordinates of the human body joints; Acquire a second angle between two connected bones of a human body in the same frame of image according to the second three-dimensional coordinates; The standard angle variation of the second angle between two adjacent frames of images is obtained.

5. A dance movement scoring device, characterized in that: include: A first recognition unit is used to input the human body posture recognition data of the current student's dance movements into the human body recognition model to obtain the first three-dimensional coordinates of the human body joints; A first acquisition unit, configured to acquire a first angle between two connected bones of a human body in a same frame of image according to the first three-dimensional coordinates; A second acquisition unit, used for acquiring a first angle change of the first angle between two adjacent frames of images; A comparison unit is used to compare the first angle variation with the standard angle variation of a standard action to obtain an angle variation difference, wherein the first angle variation matches an action sequence segment of the dance action where the corresponding standard angle variation is located; when comparing the angle variation difference, it is required to align the time of the first angle variation with the standard angle variation; A scoring unit, used for obtaining differentiated scores of different parts of the body of the current student according to the angle variation difference and the scoring rules corresponding to the current dance type, and generating a front-end display page according to the angle variation difference and the standard action data of the standard action, so as to compare and display the dance action of the current student with the standard action; The step of generating a front-end display page according to the angle variation difference and the standard action data of the standard action includes: Normalizing the angle variation difference to obtain a normalized difference; Generate a line-dot diagram using a drawing protocol according to the normalized difference and the standard action data, wherein the dots of the line-dot diagram are superimposed on corresponding positions of corresponding frames of the standard video of the standard action; The display colors of the dots and lines of the line-dot diagram are changed according to the angle variation difference and the set color rule.

6. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the steps of the method according to any one of claims 1 to 4 are implemented.

7. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 4 are implemented.