Dance posture recognition and interaction method and system based on binocular data acquisition equipment
Through binocular data acquisition equipment, synchronous recording and analysis of dance videos, identifying music beats and body key points, and calculating in-depth information, it solves the problems of incomplete evaluation and lack of data recording in traditional dance teaching, and realizes accurate evaluation of students' dance postures and feedback, improving the quality of teaching.
Patent Information
- Application Number
- CN202411499358.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-25
- Publication Date
- 2024-11-22
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
Traditional dance teaching is difficult to comprehensively and continuously evaluate and correct students' dance postures, and lacks detailed data records, which affects students' long-term progress and comparison.
The dance posture recognition method based on binocular data acquisition equipment is adopted, and dance music and student dance videos are synchronized, the music beat is detected, the key points of the limbs are extracted, the depth information is calculated, the degree of stretching and movement amplitude are judged, and the feedback report is generated.
Accurate evaluation and feedback on students' dance postures is achieved, detailed data records are provided, and teachers can quickly identify and correct problems in students' movements and improve teaching quality.
Smart Images

Figure CN119007302A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of image recognition interaction, and in particular to a dance posture recognition and interaction method and system based on a binocular data acquisition device. Background Art
[0002] In dance teaching, accurately evaluating and correcting students' dance postures is crucial to improving teaching quality. Traditional dance teaching mainly relies on teachers' visual observation and verbal guidance, which is difficult to observe in a multi-student environment and cannot be comprehensive. In addition, traditional evaluation methods lack continuity and consistency, making it difficult to provide detailed data records, which is not conducive to tracking and comparing students' long-term progress. Therefore, improvements are needed. Summary of the invention
[0003] In order to solve the above problems, the present application provides a dance posture recognition and interaction method and system based on a binocular data acquisition device.
[0004] The first object of the invention of this application is achieved through the following technical solutions: A dance posture recognition and interaction method based on binocular data acquisition equipment comprises the following steps: When receiving the collection command from the teacher, it will record the dance music and the student dance video synchronously; Detect the beat of the dance music, identify the location and time of the music, and synchronously extract the student dance video frames; Extract students' body key points from students' dance video frames; Based on the stereo matching algorithm, calculate the depth information of the key points of the student's limbs, the depth information includes the height and front and back position from the binocular data acquisition device; Based on the depth information, output the specific position of the student, wherein the specific position includes the extension degree, the movement amplitude and the spatial position; Based on the preset standard dance video, judge whether the stretching degree and movement range are standard; When it is determined that any of the stretching degree and the range of motion is not standard, focusing on the target student based on the spatial position of the target student; Generate a feedback report and send it to the teacher, the feedback report includes the non-standard student posture, specific location, and the associated student name.
[0005] In a preferred embodiment, the steps of detecting the beat of the dance music, identifying the position and time of the card points of the dance music, and synchronously extracting the student dance video frames include the steps of: Performing a preprocessing operation on the dance music, wherein the preprocessing operation includes filtering and denoising; Perform beat detection on the dance music based on the audio processing algorithm, identify and mark the card point position and time point of the dance music, wherein the card point position includes the drum beat position and the accent position; Performing a preprocessing operation on the student dance video, wherein the preprocessing operation includes denoising, contrast enhancement, and sharpening; Based on the card point position and time of the dance music, the video frames of the student dance video corresponding to the time after the preprocessing operation are extracted in real time.
[0006] In a preferred embodiment, the step of extracting key points of student limbs from student dance video frames comprises the steps of: Based on the preset posture estimation algorithm, the key points of the students' limbs are extracted and marked, including shoulders, elbows, wrists, hips, knees and ankles, and the two-dimensional coordinates of the key points of the students' limbs are output. .
[0007] In a preferred embodiment, the step of calculating the depth information of the key points of the student's limbs based on the stereo matching algorithm, wherein the depth information includes the height and front and back positions from the binocular data acquisition device, comprises the following steps: The binocular data acquisition device has two camera devices; For the same student, based on the stereo matching algorithm, the dance videos captured by the two cameras are compared, the corresponding body key points are identified and matched, and the parallax of the corresponding body key points is calculated. ; Based on preset formula , represents the distance between the two cameras, Represents the focal length of the two cameras and calculates the depth of the corresponding limb key points ; based on , output distance to the front and rear position of the binocular data acquisition device; based on , , output the 3D coordinates of the limb key points ; based on , output the height from the binocular data acquisition device.
[0008] In a preferred embodiment, the step of outputting the specific position of the student based on the depth information, wherein the specific position includes the extension degree, the movement amplitude and the spatial position, comprises the steps of: Connect the key points of the limbs in sequence to build a dance posture model; In the dance posture model, calculate the distance between adjacent limb key points , calculate the range of motion of the limb key points ; based on , output stretch degree; based on , output action amplitude; based on , output spatial position.
[0009] In a preferred embodiment, the step of judging whether the stretching degree and the movement range are standard based on the preset standard dance video comprises the steps of: Will and contrast, Represents the distance between adjacent limb key points of the preset standard dance video. , , the degree of stretching is not judged standardly; Will and contrast, Indicates the range of motion of the key points of the limbs in the preset standard dance video. , , judging the movement amplitude is not standard.
[0010] The second object of the invention is achieved by the following technical solutions: A dance posture recognition and interaction system based on binocular data acquisition equipment, comprising: The acquisition and recording module: when receiving the acquisition instruction from the teacher, it will synchronously record the dance music and the student dance video; Video frame extraction module: detect the beat of dance music, identify the card point position and time of dance music, and synchronously extract student dance video frames; Extraction module of body key points: extracting students' body key points from students' dance video frames; Depth information calculation module: Based on the stereo matching algorithm, the depth information of the key points of the student's limbs is calculated, and the depth information includes the height and front and back positions from the binocular data acquisition device; Specific position module: Based on the depth information, output the specific position of the student, which includes the extension degree, movement amplitude and spatial position; Judgment module: Based on the preset standard dance video, judge whether the stretching degree and movement range are standard; Focus module: when it is determined that any of the stretching degree and movement amplitude is not standard, focusing on the target student based on the spatial position of the target student; Feedback report module: Generates a feedback report and sends it to the teacher. The feedback report includes the non-standard student posture, specific location, and associated student names.
[0011] The third objective of the present application is achieved through the following technical solutions: A computer device comprises a memory, a processor and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the dance posture recognition and interaction method based on binocular data acquisition equipment are implemented.
[0012] The fourth objective of the present application is achieved through the following technical solutions: A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the dance posture recognition and interaction method based on a binocular data acquisition device are implemented.
[0013] In summary, the present application includes at least one of the following beneficial technical effects: 1. When receiving instructions from the teacher to collect data, it will immediately start to record two aspects of content synchronously: one is the dance music, that is, the background music used by students when dancing; the other is the student's dance video, that is, the student's dance performance. Such synchronous recording ensures the synchronization of music and dance movements, which is convenient for subsequent analysis and evaluation of students' dance performances; the rhythm analysis of the dance music detects the beat of the music, and identifies the key rhythm points and their specific time positions in the music. Then, according to these time points, the corresponding video frames are accurately extracted from the student's dance video to ensure the synchronization of dance movements and music rhythms, so as to carry out subsequent posture analysis and evaluation; in the student dance video frames that have been synchronously extracted, the key points on the student's limbs are detected and marked. These key points usually include wrists, elbows, shoulders, knees, ankles and other parts. By identifying these key points, students' dance movements can be analyzed more accurately, providing quantitative data and feedback for dance training; stereo matching algorithms are used to process images captured by binocular data acquisition devices to calculate the depth information of students' limb key points. This process includes determining the position of each key point in three-dimensional space, that is, their height and front-to-back position from the acquisition device. In this way, we can not only know how students move on a two-dimensional plane, but also understand the specific position of their movements in three-dimensional space, thereby providing more comprehensive data for the analysis and evaluation of dance movements; based on the depth information calculated previously, the specific position data of the students is output, which includes the degree of extension or bending of the students' limbs, the range of limb movement, and the specific coordinates of the students in three-dimensional space. This information helps to more accurately evaluate the quality of students' dance movements and ensure the accuracy and expressiveness of the movements; by comparing the students' dance movements with the preset standard dance videos, it is judged whether the students' extension and movement amplitude meet the prescribed standards. This process involves comparative analysis to determine the accuracy of students' movements and the quality of dance performance; when it is found that the students' extension or movement amplitude does not meet the standard, the camera focus is automatically adjusted according to the spatial position information of the target student to ensure that the target student becomes the focus of the field of view, so that the teacher can further analyze and guide the student's dance movements; after identifying that the students' extension or movement amplitude is not standard, a feedback report will be automatically generated and sent to the teacher's terminal device. This feedback report contains the following information: non-standard student postures, the specific location of the student in space, and the name of the student associated with these non-standard postures. In this way, teachers can quickly understand which students have problems with their movements and where they are, so as to provide targeted guidance and corrections.
[0014] 2. Preprocess the dance music, including filtering and denoising, to remove noise and improve the sound quality. Then, the preprocessed music is beat detected through an audio processing algorithm to identify and mark the card positions and time points in the music, such as drum beat positions and accent positions, which are crucial for the synchronization of dance movements. Preprocess the student dance video, including denoising, contrast enhancement and sharpening, to improve the video quality and make subsequent analysis more accurate. Finally, based on the card point position and time of the dance music, the video frames of the corresponding time are extracted in real time from the preprocessed student dance video to facilitate matching with the music card points, thereby evaluating whether the student's dance movements are synchronized with the music beat.
[0015] 3. Use the preset posture estimation algorithm to analyze the limbs in the student dance video, identify and mark key limb parts such as shoulders, elbows, wrists, hips, knees and ankles, and calculate the two-dimensional coordinate positions of these key points in the video screen to facilitate subsequent accurate analysis and evaluation of the student's dance movements.
[0016] 4. Compare the videos captured by the two cameras through the stereo matching algorithm, identify and match the body key points of the same student, and calculate the disparity of these key points Next, use the preset formula , combined with the distance between the two cameras and focal length , calculate the depth of the limb key points Then, the front and back positions of the students relative to the binocular data acquisition device are output, and the three-dimensional coordinates of the key points of the limbs are output. , output the height of the key points of the student's limbs from the acquisition device. This series of steps together realizes the three-dimensional spatial positioning of the student's dance movements. BRIEF DESCRIPTION OF THE DRAWINGS
[0017] Figure 1 This is a flowchart of an implementation of a dance posture recognition and interaction method embodiment based on a binocular data acquisition device of the present application; Figure 2 This is a flowchart for implementing step S20 in an embodiment of a dance posture recognition and interaction method based on a binocular data acquisition device of the present application; Figure 3 This is a flowchart for implementing step S30 in an embodiment of a dance posture recognition and interaction method based on a binocular data acquisition device of the present application; Figure 4 This is a flowchart for implementing step S40 in an embodiment of a dance posture recognition and interaction method based on a binocular data acquisition device of the present application; Figure 5This is a flowchart for implementing step S50 in an embodiment of a dance posture recognition and interaction method based on a binocular data acquisition device of the present application; Figure 6 This is a flowchart for implementing step S60 in an embodiment of a dance posture recognition and interaction method based on a binocular data acquisition device of the present application; Figure 7 This is a principle block diagram of a computer device of the present application. DETAILED DESCRIPTION
[0018] The following is combined with Figure 1-7 This application is described in further detail.
[0019] In one embodiment, if Figure 1 As shown, the present application discloses a dance posture recognition and interaction method based on a binocular data acquisition device, which specifically includes the following steps: S10: When receiving the collection instruction from the teacher, synchronously record the dance music and the student dance video; In this embodiment, the teacher end generally refers to a device or interface used to control the teaching process, which is operated by the teacher. S10 means that when receiving an instruction from the teacher end to collect data, two aspects of content will be synchronously recorded immediately: one is the dance music, that is, the background music used by the students when dancing; the other is the students' dance video, that is, the students' dance performance. Such synchronous recording ensures the synchronization of music and dance movements, which is convenient for subsequent analysis and evaluation of students' dance performance.
[0020] S20: Detect the beat of the dance music, identify the card point position and time of the dance music, and synchronously extract the student dance video frame; In this embodiment, step S20 involves rhythm analysis of the dance music, detecting the beat of the music, and identifying key rhythm points and their specific time positions in the music. Then, based on these time points, corresponding video frames are accurately extracted from the student's dance video to ensure the synchronization of dance movements with the rhythm of the music, so as to carry out subsequent posture analysis and evaluation.
[0021] S30: extracting key points of student limbs from student dance video frames; In this embodiment, step S30 refers to detecting and marking key points on the student's limbs in the synchronously extracted student dance video frames, which usually include wrists, elbows, shoulders, knees, ankles, etc. By identifying these key points, the student's dance movements can be analyzed more accurately, providing quantitative data and feedback for dance training.
[0022] S40: Calculate the depth information of the key points of the student's limbs based on a stereo matching algorithm, wherein the depth information includes the height and front and back positions from the binocular data acquisition device; In this embodiment, step S40 involves using a stereo matching algorithm to process the images captured by the binocular data acquisition device to calculate the depth information of the key points of the student's limbs. This process includes determining the position of each key point in three-dimensional space, that is, their height and front-to-back position from the acquisition device. In this way, not only can we know how the students move on a two-dimensional plane, but also understand the specific position of their movements in three-dimensional space, thereby providing more comprehensive data for the analysis and evaluation of dance movements.
[0023] S50: Based on the depth information, output the specific position of the student, wherein the specific position includes the stretching degree, the movement amplitude and the spatial position; In this embodiment, the extension degree refers to the degree to which the student's limbs are stretched or bent, the movement amplitude refers to the range of movement of the limbs, and the spatial position refers to the specific coordinates of the student in three-dimensional space; Specifically, step S50 outputs the student's specific position data based on the depth information calculated previously, including the degree of extension or bending of the student's limbs, the range of limb movement, and the student's specific coordinates in three-dimensional space. This information helps to more accurately evaluate the quality of the student's dance movements and ensure the accuracy and expressiveness of the movements.
[0024] S60: Based on the preset standard dance video, determine whether the stretching degree and the movement range are standard; In this embodiment, step S60 is to compare the student's dance movements with a preset standard dance video to determine whether the student's stretching degree and movement range meet the prescribed standards. This process involves comparative analysis to determine the accuracy of the student's movements and the quality of the dance performance.
[0025] S70: When it is determined that any of the stretching degree and the movement amplitude is not standard, focusing on the target student based on the spatial position of the target student; In this embodiment, step S70 is to automatically adjust the camera focus according to the spatial position information of the target student when it is found that the student's stretching degree or movement amplitude does not meet the standard, to ensure that the target student becomes the focus of the field of view, so as to facilitate the teacher to further analyze and guide the student's dance movements.
[0026] S80: Generate a feedback report and send it to the teacher, where the feedback report includes the non-standard student posture, specific location, and associated student name.
[0027] In this embodiment, step S80 means that after identifying that the student's stretching degree or movement range is not standard, a feedback report will be automatically generated and sent to the teacher's terminal device. This feedback report contains the following information: the non-standard student posture, the specific location of the student in space, and the name of the student associated with these non-standard postures. In this way, the teacher can quickly understand which students have problems with their movements and their locations, so as to provide targeted guidance and correction.
[0028] Reference Figure 2 , step S20, comprising the steps of: S201: performing a preprocessing operation on the dance music, wherein the preprocessing operation includes filtering and denoising; S202: Performing beat detection on the dance music based on an audio processing algorithm, identifying and marking the card point positions and time points of the dance music, wherein the card point positions include drum beat positions and accent positions; S203: performing a preprocessing operation on the student dance video, wherein the preprocessing operation includes denoising, contrast enhancement, and sharpening; S204: Based on the card point position and time of the dance music, the video frames of the student dance video corresponding to the time after the pre-processing operation are extracted in real time.
[0029] In the present embodiment, steps S201 to S204 involve processing of accompanying dance music and student dance videos so as to more accurately analyze and evaluate the student's dance performance. In step S201, the accompanying dance music is pre-processed, including filtering and denoising, to remove noise in the music and improve the sound quality. Then, step S202 performs beat detection on the pre-processed music through an audio processing algorithm, identifies and marks the card point position and time point in the music, such as the drum position and the accent position, which are crucial for the synchronization of dance movements. In step S203, the student dance video is pre-processed, including denoising, contrast enhancement and sharpening, to improve the video quality, so that subsequent analysis is more accurate. Finally, in step S204, based on the card point position and time of the accompanying dance music, the video frame of the corresponding time is extracted in real time from the pre-processed student dance video, so as to match with the music card point, thereby evaluating whether the student's dance movements are synchronized with the music beat.
[0030] Reference Figure 3 , step S30, comprising the steps of: S301: Based on the preset posture estimation algorithm, extract and mark the key points of the student's limbs, including shoulders, elbows, wrists, hips, knees and ankles, and output the two-dimensional coordinates of the key points of the student's limbs .
[0031] In this embodiment, in step S301, a preset posture estimation algorithm is used to analyze the limbs in the student dance video, identify and mark key limb parts such as shoulders, elbows, wrists, hips, knees and ankles, and calculate the two-dimensional coordinate positions of these key points in the video screen, so as to facilitate the subsequent accurate analysis and evaluation of the student's dance movements.
[0032] Reference Figure 4 , step S40, comprising the steps of: S401: The binocular data acquisition device has two camera devices; S402: For the same student, based on the stereo matching algorithm, compare the dance videos captured by the two cameras, identify and match the corresponding body key points, and calculate the parallax of the corresponding body key points ; S403: Based on the preset formula , represents the distance between the two cameras, Represents the focal length of the two cameras and calculates the depth of the corresponding limb key points ; S404: Based on , output distance to the front and rear position of the binocular data acquisition device; S405: Based on , , output the 3D coordinates of the limb key points ; S406: Based on , output the height from the binocular data acquisition device.
[0033] In this embodiment, in steps S401 to S406, the binocular data acquisition device uses two cameras to synchronously capture the student's dance video. S402 compares the videos captured by the two cameras through a stereo matching algorithm, identifies and matches the body key points of the same student, and calculates the parallax of these key points. Then, S403 uses the preset formula , combined with the distance between the two cameras and focal length , calculate the depth of the limb key points Then, S404 outputs the front and rear positions of the student relative to the binocular data acquisition device, and S405 outputs the three-dimensional coordinates of the key points of the limbs. , and S406 outputs the height of the key points of the student's limbs from the acquisition device. This series of steps together realizes the three-dimensional spatial positioning of the student's dance movements.
[0034] Reference Figure 5 , step S50, comprising the steps of: S501: Connect the key points of the limbs in sequence to construct a dance posture model; S502: In the dance posture model, calculate the distance between adjacent limb key points , calculate the range of motion of the limb key points ; S503: Based on , output stretch degree; S504: Based on , output action amplitude; S505: Based on , output spatial position.
[0035] In this embodiment, steps S501 to S505 involve constructing and evaluating a dance posture model. In step S501, the identified limb key points are sequentially connected to form a dance posture model, which helps to visualize the student's dance movements. Next, step S502 calculates the distance between adjacent limb key points in the model. , and the range of motion of key limb points , which provides quantitative data for evaluating dance movements. Then, based on these calculations, step S503 outputs the student's stretching degree, step S504 outputs the movement amplitude, and step S505 outputs the student's specific position in three-dimensional space, which together constitute a comprehensive evaluation of the student's dance movements.
[0036] Reference Figure 6 , step S60, comprising the steps of: S601: and contrast, Represents the distance between adjacent limb key points of the preset standard dance video. , , the degree of stretching is not judged standardly; S602: and contrast, Indicates the range of motion of the key points of the limbs in the preset standard dance video. , , judging the movement amplitude is not standard.
[0037] In this embodiment, steps S601 and S602 are used to analyze and compare the student's dance movements with the movement parameters in the standard dance video. In step S601, the distances between adjacent limb key points in the student's dance video are measured and compared. The distance from the corresponding key point in the preset standard dance video .when , In step S602, the range of motion of the key points of the limbs in the student's dance video is calculated and compared. Range of motion with preset standard dance videos .when , , it will be judged that the student's movements are not standard. These two steps automatically identify inaccuracies or inadequacies in students' dance movements by setting specific evaluation criteria.
[0038] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.
[0039] In one embodiment, a dance posture recognition and interaction system based on a binocular data acquisition device is provided, and the dance posture recognition and interaction system based on a binocular data acquisition device corresponds to the dance posture recognition and interaction method based on a binocular data acquisition device in the above embodiment. The dance posture recognition and interaction system based on a binocular data acquisition device includes: Acquisition and recording module: when receiving the acquisition instruction from the teacher, it will synchronously record the dance music and student dance video; Video frame extraction module: detect the beat of dance music, identify the card point position and time of dance music, and synchronously extract student dance video frames; Extraction module of body key points: extracting students' body key points from students' dance video frames; Depth information calculation module: Based on the stereo matching algorithm, the depth information of the key points of the student's limbs is calculated, and the depth information includes the height and front and back positions from the binocular data acquisition device; Specific position module: Based on the depth information, output the specific position of the student, which includes the extension degree, movement amplitude and spatial position; Judgment module: Based on the preset standard dance video, judge whether the stretching degree and movement range are standard; Focus module: when it is determined that any of the stretching degree and movement amplitude is not standard, focusing on the target student based on the spatial position of the target student; Feedback report module: Generates a feedback report and sends it to the teacher. The feedback report includes the non-standard student posture, specific location, and associated student names.
[0040] Optionally, also include: The first pre-processing operation module is used to perform pre-processing operations on the dance music, wherein the pre-processing operations include filtering and denoising; Beat detection module: based on the audio processing algorithm, the beat detection of the dance music is performed to identify and mark the card point position and time point of the dance music, where the card point position includes the drum point position and the accent position; The second preprocessing operation module is used to perform preprocessing operations on the student dance video, wherein the preprocessing operations include denoising, contrast enhancement, and sharpening; The first extraction module: based on the card point position and time of the dance music, the video frames of the student dance video corresponding to the time after the pre-processing operation are extracted in real time.
[0041] Optionally, also include: The second extraction module: based on the preset posture estimation algorithm, extract and mark the key points of the student's limbs, including shoulders, elbows, wrists, hips, knees and ankles, and output the two-dimensional coordinates of the key points of the student's limbs .
[0042] Optionally, also include: Binocular camera module: The binocular data acquisition device has two cameras; Recognition and matching module: For the same student, based on the stereo matching algorithm, compare the dance videos collected by the two cameras, identify and match the corresponding limb key points, and calculate the disparity of the corresponding limb key points ; Calculation depth module: based on preset formula , represents the distance between the two cameras, Represents the focal length of the two cameras and calculates the depth of the corresponding limb key points ; Front and rear position module: based on , output distance to the front and rear position of the binocular data acquisition device; Three-dimensional coordinate module: based on , , output the 3D coordinates of the limb key points ; Height module: Based on , output the height from the binocular data acquisition device.
[0043] Optionally, also include: Dance posture module: connect the key points of the limbs in sequence to build a dance posture model; Calculation module: Calculate the distance between adjacent limb key points in the dance posture model , calculate the range of motion of the limb key points ; Stretch degree module: based on , output stretch degree; Range of motion module: based on , output action amplitude; Spatial location module: based on , output spatial position.
[0044] Optionally, also include: Determine the stretch degree module: and contrast, Represents the distance between adjacent limb key points of the preset standard dance video. , , the degree of stretching is not judged standardly; Action range judgment module: and contrast, Indicates the range of motion of the key points of the limbs in the preset standard dance video. , , judging the movement amplitude is not standard.
[0045] For the specific definition of a dance posture recognition and interaction system based on binocular data acquisition equipment, please refer to the definition of a dance posture recognition and interaction method based on binocular data acquisition equipment in the above text, which will not be repeated here. Each module in the above dance posture recognition and interaction system based on binocular data acquisition equipment can be implemented in whole or in part by software, hardware and a combination thereof. The above modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory in the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.
[0046] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 7 As shown. The computer device includes a processor, a memory, a network interface and a database connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store limb key points, the distance between adjacent limb key points, and the range of motion of limb key points. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, it is a dance posture recognition and interaction method based on a binocular data acquisition device.
[0047] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, a dance posture recognition and interaction method based on a binocular data acquisition device is implemented.
[0048] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, a dance posture recognition and interaction method based on a binocular data acquisition device is implemented.
[0049] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.
[0050] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.
Claims
1. A dance posture recognition and interaction method based on binocular data acquisition equipment, characterized in that: Includes steps: When receiving the collection command from the teacher, it will record the dance music and the student dance video synchronously; Detect the beat of the dance music, identify the location and time of the music, and synchronously extract the student dance video frames; Extract students' body key points from students' dance video frames; Based on the stereo matching algorithm, calculate the depth information of the key points of the student's limbs, the depth information includes the height and front and back position from the binocular data acquisition device; Based on the depth information, output the specific position of the student, wherein the specific position includes the extension degree, the movement amplitude and the spatial position; Based on the preset standard dance video, judge whether the stretching degree and movement range are standard; When it is determined that any of the stretching degree and the range of motion is not standard, focusing on the target student based on the spatial position of the target student; Generate a feedback report and send it to the teacher, the feedback report includes the non-standard student posture, specific location, and the associated student name.
2. The dance posture recognition and interaction method based on binocular data acquisition equipment according to claim 1 is characterized in that: The steps of detecting the beat of the dance music, identifying the position and time of the card points of the dance music, and synchronously extracting the student dance video frames include the following steps: Performing a preprocessing operation on the dance music, wherein the preprocessing operation includes filtering and denoising; Perform beat detection on the dance music based on the audio processing algorithm, identify and mark the card point position and time point of the dance music, wherein the card point position includes the drum beat position and the accent position; Performing a preprocessing operation on the student dance video, wherein the preprocessing operation includes denoising, contrast enhancement, and sharpening; Based on the card point position and time of the dance music, the video frames of the student dance video corresponding to the time after the preprocessing operation are extracted in real time.
3. The dance posture recognition and interaction method based on binocular data acquisition equipment according to claim 1 is characterized in that: The step of extracting key points of student limbs from student dance video frames comprises the steps of: Based on the preset posture estimation algorithm, the key points of the students' limbs are extracted and marked, including shoulders, elbows, wrists, hips, knees and ankles, and the two-dimensional coordinates of the key points of the students' limbs are output. .
4. The dance posture recognition and interaction method based on binocular data acquisition equipment according to claim 1 is characterized in that: The step of calculating the depth information of the key points of the student's limbs based on the stereo matching algorithm, wherein the depth information includes the height and the front and back position from the binocular data acquisition device, comprises the following steps: The binocular data acquisition device has two camera devices; For the same student, based on the stereo matching algorithm, the dance videos captured by the two cameras are compared, the corresponding body key points are identified and matched, and the parallax of the corresponding body key points is calculated. ; Based on preset formula , represents the distance between the two cameras, Represents the focal length of the two cameras and calculates the depth of the corresponding limb key points ; based on , output distance to the front and rear position of the binocular data acquisition device; based on , , output the 3D coordinates of the limb key points ; based on , output the height from the binocular data acquisition device.
5. The dance posture recognition and interaction method based on binocular data acquisition equipment according to claim 1 is characterized in that: The step of outputting the specific position of the student based on the depth information, wherein the specific position includes the extension degree, the movement amplitude and the spatial position, comprises the steps of: Connect the key points of the limbs in sequence to build a dance posture model; In the dance posture model, calculate the distance between adjacent limb key points , calculate the range of motion of the limb key points ; based on , output stretch degree; based on , output action amplitude; based on , output spatial position.
6. The dance posture recognition and interaction method based on binocular data acquisition equipment according to claim 1 is characterized in that: The step of judging whether the stretching degree and the movement range are standard based on the preset standard dance video comprises the following steps: Will and contrast, Represents the distance between adjacent limb key points of the preset standard dance video. , , the degree of stretching is not judged standardly; Will and contrast, Indicates the range of motion of the key points of the limbs in the preset standard dance video. , , judging the movement amplitude is not standard.
7. A dance posture recognition and interaction system based on binocular data acquisition equipment, characterized in that: include: Acquisition and recording module: when receiving the acquisition instruction from the teacher, it will synchronously record the dance music and student dance video; Video frame extraction module: detect the beat of dance music, identify the card point position and time of dance music, and synchronously extract student dance video frames; Extraction module of body key points: extracting students' body key points from students' dance video frames; Depth information calculation module: Based on the stereo matching algorithm, the depth information of the key points of the student's limbs is calculated, and the depth information includes the height and front and back positions from the binocular data acquisition device; Specific position module: Based on the depth information, output the specific position of the student, which includes the extension degree, movement amplitude and spatial position; Judgment module: Based on the preset standard dance video, judge whether the stretching degree and movement range are standard; Focus module: when it is determined that any of the stretching degree and movement amplitude is not standard, focusing on the target student based on the spatial position of the target student; Feedback report module: Generates a feedback report and sends it to the teacher. The feedback report includes the non-standard student posture, specific location, and associated student names.
8. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of a dance posture recognition and interaction method based on a binocular data acquisition device as described in any one of claims 1 to 6 when executing the computer program.
9. A computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the steps of a dance posture recognition and interaction method based on a binocular data acquisition device as described in any one of claims 1 to 6.
Citation Information
Patent Citations
Video recording method, photographing device, electronic device and medium
CN107682654A
Dance posture feedback method and device
CN112416124A
Group queue training analysis method and device
CN114093029A
Human body posture high-precision identification method based on binocular vision
CN115578454A
Cited By
Dance posture action recognition and interaction method and system based on visual calculation
CN119600691A
A dance posture recognition and interaction method and system based on visual computing
CN119600691B
Artificial intelligence-based dance posture correction method and system
CN121053698A
Human body action evaluation method and system based on camera recognition
CN121482873A
Human action evaluation method and system based on camera recognition
CN121482873B
