Interactive online dance teaching method and device

Abstract

The present invention relates to an interactive online dance teaching method and device, belonging to the field of interactive television communication technology. The method is based on a visualization terminal capable of running an interactive dance teaching system, wherein the interactive dance teaching system includes an acquisition module and a reminder module. The acquisition module includes an image acquisition unit, a tactile feedback unit, and a processing unit. The image acquisition unit is used to capture dynamic images of the user's posture in real time. The method uses the visualization terminal to capture the user's real-time dance posture, dance movement by dance movement, and extracts key points of the user's limbs from dance images. By comparing the user's limb positions with key points of the limbs in a preset dance teaching video and providing intuitive feedback, the user can promptly identify and adjust their posture errors, thereby improving learning efficiency and effectiveness through such visual interaction.

Description

Technical Field

[0001] The present invention belongs to the technical field of interactive television communications, and in particular relates to an interactive online dance teaching method and device. Background Art

[0002] With the development of technology, the learning of various skills will be transplanted into videos and played using mobile phones, TVs or computers. Because of its repeated learning and relatively low learning costs, it has gained the preference of many people, and teaching is carried out through live broadcasts or rebroadcasts.

[0003] This learning model is especially suitable for learning in the field of dance. However, the existing video dance teaching does not have the function of correcting students' postures. The process of learning dance is to constantly adjust one's own posture to make one's movements more standard. However, blindly watching teaching videos does not have a teaching purpose similar to interactive teaching between teachers and students, so it is difficult to achieve good learning results.

[0004] Therefore, an interactive online dance teaching method and device are proposed to solve the above problems. Summary of the Invention

[0005] Technical problems solved

[0006] In view of the above-mentioned shortcomings of the prior art, the present invention provides an interactive online dance teaching method and device, which can effectively solve the problem in the prior art that the learning process of learning dance through teaching videos lacks interaction and leads to poor learning effect.

[0007] Technical Solution

[0008] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0009] The present invention provides an interactive online dance teaching method, which is based on a visual terminal capable of running an interactive dance teaching system, wherein the interactive dance teaching system includes a collection module and a reminder module;

[0010] The acquisition module includes an image acquisition unit, a tactile feedback unit and a processing unit;

[0011] The image acquisition unit is used to acquire dynamic images of the user's posture in real time;

[0012] The tactile feedback unit determines the user's standing position based on the user's real-time standing position, and is located on the ground and in front of the image acquisition unit. It includes a 3x3 matrix grid, and each matrix grid is provided with regularly arranged pressure sensing sub-units;

[0013] The processing unit determines the user's real-time dance posture information based on the collected dynamic posture image and standing position, and sends the real-time dance posture information to the visualization terminal. The visualization terminal creates a dynamic virtual user image based on the user's real-time dance posture information and compares the real-time dance posture information with preset dance video information.

[0014] The dance video information is displayed in the form of a virtual dancer's dance posture. The visualization terminal generates a reminder signal based on the comparison result between the real-time dance posture information and the preset dance video information. The reminder module sends an instruction to the user according to the reminder signal.

[0015] The comparison result includes the user's incorrect posture interval information and incorrect standing position information.

[0016] Furthermore, the steps for establishing the dynamic virtual user image are as follows:

[0017] S1: First, collect the user's dance video within a certain time interval and identify the user's body information in each frame of the video;

[0018] S2: Secondly, the user's foot posture and standing position are determined by the number and distribution of the pressure sensing sub-units triggered in the tactile feedback unit;

[0019] S3: The visualization terminal extracts the user's limb information in step S1 and the user's foot posture and standing position in step S2 to generate a virtual user image, and accelerates multiple consecutive frames of the virtual user image to form a video;

[0020] The user's limb information includes the positions of the user's hands and feet.

[0021] Furthermore, the steps of comparing the real-time dance posture information with the preset dance video information are as follows:

[0022] T1: Create a three-dimensional coordinate system with the center of the matrix in the tactile feedback unit as the origin, the length and width of the matrix as the x-axis and y-axis respectively, and the user's height as the z-axis. Mark the user's limb information in step S1 in the three-dimensional coordinate system.

[0023] T2: Divide the preset dance video information into dance pictures frame by frame according to the time interval in step S1, and extract the key points of the limb positions in the dance pictures and mark them in the three-dimensional coordinate system in step T1;

[0024] T3: Compare the marking points in steps T1 and T2. If there is a deviation between the user's limb information and orientation marks and the limb position key points in the preset video information, it is considered as a non-standard dance posture and counted;

[0025] The preset dance video information is any frame image in the dance video reserved in the visualization terminal, and the extracted user limb information and limb position key points are all captured from the same time node and are all located in the same complete dance action.

[0026] Furthermore, the calculation method of the incorrect posture interval information is as follows:

[0027] T301: Let any coordinate point in the user's limb information in step T1 be m, and the coordinates of m are (x0, y0, z0). Let the part of the limb position key point corresponding to m in step T2 be n, and the coordinates of n are (x1, y1, z1).

[0028] T302: Set the deviation threshold length to l;

[0029] T303: When ≤l, ≤l and When ≤l are satisfied, it is regarded as a standard posture; otherwise it is regarded as a non-standard posture and counted, denoted as p (p is an integer greater than 0).

[0030] Furthermore, the visualization terminal further includes a counting unit for determining whether any complete dance movement is standard. The specific determination steps are as follows:

[0031] Step 1: Based on the same complete dance action, extract video clips from the real-time dance posture information and dance video information respectively;

[0032] Step 2: Select several image segments at the same time point from the real-time dance posture information video segment and the dance video information video segment obtained in step 1;

[0033] Step 3: Set the number of non-standard postures allowed in a complete dance move to q (q is an integer greater than 0);

[0034] Step 4: Compare the number of times p appears in the image segments obtained in step 2 after calculating the wrong posture interval information with q;

[0035] Among them, when If the dance moves meet the standards, the dance moves are deemed to meet the standards; otherwise, the dance moves are deemed to meet the standards.

[0036] Furthermore, the step T303 further includes:

[0037] T304: Set the gradient comparison threshold, respectively 、 and ;

[0038] T304: Compare the user's limb information in step T303 with the deviation values ​​of the key position points of the limbs, and score them respectively;

[0039] when ≤ 3 points;

[0040] when ≤ 3 points;

[0041] when ≤ 3 points,

[0042] when ≤ 2 points;

[0043] when ≤ 2 points;

[0044] when ≤ 2 points;

[0045] when ≤ 1 point;

[0046] when ≤ 1 point;

[0047] when ≤ 1 point;

[0048] when 、 and If the value of is greater than 1, no points will be scored;

[0049] T305: Count the scores of each complete dance move in the entire dance video information.

[0050] Furthermore, the visualization terminal may display a dynamic virtual user image created based on the user's real-time dance posture information and a virtual dancer image on the same screen, or blur the virtual dancer image and display it with the dynamic virtual user image.

[0051] Furthermore, the visualization terminal may adjust the size of the virtual dancer's image in proportion to the user's dynamic virtual user image.

[0052] Furthermore, the indication information includes at least one or more of voice reminder, visual terminal reminder, and light conversion.

[0053] The present invention also provides an interactive online dance teaching device, comprising a display screen (1) and a dance mat (2), wherein the display screen (1) is provided with a speaker, and has a built-in memory and a controller, wherein the memory stores dance video information of a plurality of preset dances, and the controller has a built-in program capable of running the interactive dance teaching method described in claims 1-9, and the dance mat (2) is provided at the front end of the display screen (1), and has a plurality of pressure sensors regularly distributed therein that are in communication with the controller.

[0054] Beneficial effects

[0055] Compared with the known public technologies, the technical solution provided by the present invention has the following beneficial effects:

[0056] The present invention collects the user's real-time dance posture one dance movement at a time through a visualization terminal, and extracts the key points of the user's limbs from the dance picture. By comparing the user's limb positions with the key points of the limbs in a preset dance teaching video and providing intuitive feedback, the user can promptly discover his or her own posture errors and make adjustments, thereby improving learning efficiency and effectiveness through such visual interaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] To more clearly illustrate the embodiments of the present invention or the technical solutions in 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 only some embodiments of the present invention. Those skilled in the art can also derive other drawings based on these drawings without inventive effort.

[0058] Figure 1 This is an example diagram of the steps of the interactive online dance teaching method in an embodiment of the present invention;

[0059] Figure 2 A schematic diagram of the process of an interactive online dance teaching system in an embodiment of the present invention;

[0060] Figure 3 This is a schematic diagram of a display screen and a dance mat in an embodiment of the present invention;

[0061] Figure 4 This is a schematic diagram of judging dance movements that meet the standards in an embodiment of the present invention;

[0062] Figure 5 Schematic diagram of judging dance movements that do not meet the standards in an embodiment of the present invention.

[0063] Figure numerals: 1. display screen; 2. dance mat. DETAILED DESCRIPTION

[0064] To make the purpose, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only 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 making any creative efforts shall fall within the scope of protection of the present invention.

[0065] The present invention will be further described below with reference to the embodiments.

[0066] As short video live streaming becomes more and more popular, more and more people tend to learn dance by watching instructional videos, most of which are learning while watching on TV or mobile phones.

[0067] The beauty of dance is not only reflected in the coordination and continuity of movements, but the most important thing is the angle of the limbs. When the limbs deviate too much from the expected position in a certain movement, it will destroy the continuity and beauty of the entire dance movement. Therefore, in addition to the body movements and the music that coordinates with the body movements, in order to complete an elegant and beautiful dance posture, the optimal position of the limbs at specific time points is crucial. Such training can only be achieved through repeated practice to achieve a certain level of muscle memory.

[0068] However, in the prior art, the training video can only be paused and watched repeatedly, and long-term frequent watching will seriously affect the dancers' learning enthusiasm and learning efficiency.

[0069] To this end, after research, our company has proposed an interactive online dance teaching method. In addition to being able to play preset dance teaching videos, it can also judge the user's dance posture and position through visual acquisition + body sensing acquisition, and then compare the judgment results with the preset dance teaching video (same time node). Through corresponding reminders and scoring, users can know in real time the incorrect angles (orientations) of their limbs and the time nodes when the errors occurred, and strengthen training accordingly.

[0070] In addition, the scoring system in this solution can make the dance learning process challenging and interesting, and improve users' learning interest and enthusiasm.

[0071] Specifically, this solution is described in detail by the following examples.

[0072] Example: Refer to the attached Figure 1-5First of all, an interactive online dance teaching method in this scheme is based on a visualization terminal that can run an interactive dance teaching system. The visualization terminal has built-in software that can run an interactive dance teaching system. The interactive dance teaching system in this scheme mainly includes two modules: an acquisition module and a reminder module.

[0073] First, let me briefly introduce the general operating steps of this solution. When the user stands in front of the visualization terminal, he or she first needs to make a virtual connection and correct his or her posture. Specifically, the user's height will be collected, and a dynamic virtual user image will be established on the screen of the visualization terminal. Based on this virtual user image, the customer can adjust his or her position on the tactile feedback unit (dance mat 2). When everything is ready, the dance playback is ready to start.

[0074] The preset videos in this solution can be pre-marked according to different dance moves, mainly marking the time nodes before and after a single dance move. When the user reaches this time node, the real-time image of the user performing this dance move is collected. Based on the judgment of the image and the image of the preset video at the same time node, the error points and causes of the user's current dance move can be determined. For example, when the wrist needs to be extended to 90 degrees with the body, but the user only extends it to 70 degrees, the visualization terminal will show where the user's wrist should be placed.

[0075] The visualization terminal can display the dynamic virtual user image based on the user's real-time dance posture information and the virtual dancer image on the same screen, or blur the virtual dancer image and display it with the dynamic virtual user image.

[0076] Specifically, this solution will first generate an image of a virtual dancer (which can be a real person) in a preset dance video, and display the user's virtual user image and the virtual dancer on the same screen or in split screens. After collecting the user's height in the previous step, the height of the virtual dancer can be adjusted proportionally. Optionally, the image of the virtual dancer can be blurred and overlapped with the virtual user image (post-positioned), so that the user can observe and compare his or her own posture in real time while dancing.

[0077] The steps for establishing a dynamic virtual user image in this solution are as follows:

[0078] S1: First, collect the user's dance video within a certain time interval and identify the user's body information in each frame of the video;

[0079] S2: Secondly, the user's foot posture and standing position are determined by the number and distribution of the pressure sensing sub-units triggered in the tactile feedback unit;

[0080] S3: The visualization terminal extracts the user's limb information in step S1 and the user's foot posture and standing position in step S2 to generate a virtual user image, and accelerates multiple consecutive frames of the virtual user image to form a video;

[0081] The user's limb information includes the positions of the user's hands and feet.

[0082] It is worth noting that a virtual image of the dance mat 2 can be set to match the user's proportions on the visualization terminal. The virtual user image will step on the virtual image of the dance mat 2. When the user steps on the wrong position in a certain dance move, the dance mat 2 will display a corresponding color to remind the user that the stepping direction is wrong. The dance mat 2 in this solution, that is, the tactile feedback unit is divided into a 3x3 matrix grid, and each matrix grid is distributed with a number of pressure sensing sub-units. According to the triggering status of the pressure sensing sub-unit, the standing position of the user's two feet at this time can be judged. In practice, the dance mat 2 can be displayed on the screen of the visualization terminal in a square frame diagram and located on one side of the virtual user image, or the nine different color areas in the matrix grid can be displayed at the bottom of the virtual user image. When the user steps on the wrong position, the user can be reminded in different colors.

[0083] Specifically, the acquisition module in this solution includes an image acquisition unit, a tactile feedback unit, and a processing unit;

[0084] The image acquisition unit is used to capture the user's posture dynamic image in real time. After the acquisition, multiple dance moves will be extracted from a complete dance according to a preset dance video;

[0085] The tactile feedback unit determines the user's standing position based on the user's real-time standing position, and is located on the ground and in front of the image acquisition unit. The tactile feedback unit includes a 3x3 matrix grid, and each matrix grid is provided with regularly arranged pressure sensing sub-units;

[0086] The processing unit determines the user's real-time dance posture information based on the collected dynamic posture images and standing position, and sends the real-time dance posture information to the visualization terminal. The visualization terminal establishes a dynamic virtual user image based on the user's real-time dance posture information, and compares the real-time dance posture information with the preset dance video information.

[0087] For example, when starting a certain dance move, the entire move has 10 seconds, so any dance pictures can be extracted from the entire move. For example, with an interval of 2 seconds, 5 dance pictures will be extracted from the entire dance move (which can be edited in advance into the preset dance video information). When the user starts to dance this dance move, at the time interval of 2 seconds, the user will start to collect images, and the collected images will be analyzed and compared with the preset dance pictures to determine the deficiencies and defects of the user when dancing this move.

[0088] In general, when collecting information from the user, information will also be collected from the tactile feedback unit under the user's feet. The user's standing posture will be combined with a comparison analysis of dance pictures to determine whether the user's movements are standard.

[0089] The specific steps for comparing real-time dance posture information with preset dance video information are as follows:

[0090] T1: Create a three-dimensional coordinate system with the center of the matrix in the tactile feedback unit as the origin, the length and width of the matrix as the x-axis and y-axis respectively, and the user's height as the z-axis. Mark the user's limb information in step S1 in the three-dimensional coordinate system.

[0091] T2: Divide the preset dance video information into dance pictures frame by frame according to the time interval in step S1, and extract the key points of the limb positions in the dance pictures and mark them in the three-dimensional coordinate system in step T1;

[0092] T3: Compare the marking points in steps T1 and T2. If there is a deviation between the user's limb information and orientation marks and the limb position key points in the preset video information, it is considered as a non-standard dance posture and counted;

[0093] The preset dance video information (i.e., the dance picture captured from the preset dance video in the previous article) is any frame image in the dance video reserved in the visualization terminal. At the same time, the extracted user limb information and limb position key points are all captured from the same time node and are all located in the same complete dance movement.

[0094] It is worth noting that determining the user's limb information through images is a well-known visual detection technology, and this embodiment does not illustrate the detection process in detail.

[0095] The dance video information is displayed in the form of a virtual dancer's dance posture. The visualization terminal generates a reminder signal based on the comparison result between the real-time dance posture information and the preset dance video information. The reminder module sends an instruction to the user based on the reminder signal. The comparison result includes the user's incorrect posture interval information and incorrect standing position information, which are judged respectively through the judgment of the dance picture and the tactile feedback result of the tactile feedback unit.

[0096] The process of judging whether the user's posture is standard or not is as follows, that is, the calculation method for the incorrect posture interval information is as follows:

[0097] T301: Let any coordinate point in the user's limb information in step T1 be m, and the coordinates of m are (x0, y0, z0). Let the part of the limb position key point corresponding to m in step T2 be n, and the coordinates of n are (x1, y1, z1).

[0098] T302: Set the deviation threshold length to l;

[0099] T303: When ≤l, ≤l and If ≤l are satisfied, it is regarded as a standard posture, otherwise it is regarded as a non-standard posture and counted, recorded as p (p is an integer greater than 0), refer to the attached Figure 4 and attached Figure 5 In the figure, the distance between point m and point n can be calculated by the triangle principle. When this distance is greater than l (the radius of the sphere), the user's action is too large and does not meet the standard.

[0100] By marking the user's limb information in a three-dimensional coordinate system, and within a certain range, a small deviation between the user's posture and the preset dance video information is allowed. For example, when the arm swings forward, a small deviation within the spherical range around the arm does not affect the beauty of the dance. Therefore, by setting a threshold, the threshold inclusion range is determined based on the preset dance video information. For example, a spherical range is made with the palm part of the preset dance video information as the base point and the deviation threshold length l as its radius. When the user's palm falls within this spherical range, it is considered that the action meets the standard.

[0101] Therefore, the distance between the center of the sphere and the user's palm collection point is calculated in the three-dimensional coordinate system to determine whether it exceeds the radius of the sphere. When it exceeds this radius, that is, is greater than this threshold, it is considered that the user's action is too large and is judged as an incorrect action.

[0102] Optionally, this solution can also give a special reminder to the user when there are too many incorrect postures in a dance move. During actual use, the customer can choose to perform the dance move again, specifically by judging the number of times the dance pictures collected within the dance move have excessive deviations.

[0103] Specifically, the visualization terminal further includes a counting unit for determining whether any complete dance movement is standard. The specific determination steps are as follows:

[0104] Step 1: Based on the same complete dance action, extract video clips from the real-time dance posture information and dance video information respectively;

[0105] Step 2: Select several image segments at the same time point from the real-time dance posture information video segment and the dance video information video segment obtained in step 1;

[0106] Step 3: Set the number of non-standard postures allowed in a complete dance move to q (q is an integer greater than 0);

[0107] Step 4: Compare the number of times p appears in the image segments obtained in step 2 after calculating the wrong posture interval information with q;

[0108] when If the dance moves are correct, the dance is deemed to meet the standards. Otherwise, the dance moves are deemed to not meet the standards. The dance can be performed again.

[0109] In order to increase the competitiveness and fun of users in the dance learning process, this solution also provides a scoring method. In practical applications, such a scoring model can be used to unlock staged dance training, increasing users' interest by challenging higher difficulty levels and judging their own dance level.

[0110] Specifically, the following steps may be included after step T303 in the above-mentioned method for calculating the incorrect posture interval information:

[0111] T304: Set the gradient comparison threshold, respectively 、 and ;

[0112] T304: Compare the user's limb information in step T303 with the deviation values ​​of the key position points of the limbs, and score them respectively;

[0113] when ≤ 3 points;

[0114] when ≤ 3 points;

[0115] when ≤ 3 points,

[0116] when ≤ 2 points;

[0117] when ≤ 2 points;

[0118] when ≤ 2 points;

[0119] when ≤ 1 point;

[0120] when ≤ 1 point;

[0121] when ≤ 1 point;

[0122] when 、 and If the value of is greater than 1, no points will be scored;

[0123] T305: Count the scores of each complete dance move in the entire dance video information.

[0124] For example, when performing a segment of a certain dance movement, the user's palm collection points are scored from three directions, namely the plane coordinates of the x-axis and y-axis planes, the x-axis and z-axis planes, and the y-axis and z-axis planes. When the deviation threshold length is set to 1, this deviation threshold is graded. The closer the distance, the closer the user's posture at this time is to the posture in the preset dance video information, and the higher the score. For example, a certain dance is divided into n movements, each movement has a full score of 9n points, and 60% of the full score is used as the passing line. The dance is considered to be substandard if it is below the passing line.

[0125] Therefore, when measuring a user collection point, the distance is measured from the coordinates in the three planes and the reference point in the preset video information, and the scores in the three planes are added together. When the distance of the collection point of any plane exceeds the distance length l, no score is given for that item. The scores of the other two planes are still calculated based on the judgment of the collection point.

[0126] The visualization terminal can adjust the size of the virtual dancer's image in proportion to the user's dynamic virtual user image.

[0127] In this solution, no matter if an error occurs or the entire dance movement is unqualified or the entire dance score is unqualified, appropriate reminder information will be used to remind the user, and the instruction information includes at least one or more of voice reminders, visual terminal reminders, and light conversions.

[0128] For example, if the user's initial standing posture is not standard, the visual terminal will show the direction and distance in which the user should move. If the standing position is wrong, the tactile feedback unit can be set with a certain light module to emit colored light in the correct matrix to guide the user to stand in the correct area.

[0129] When the user's dance moves are not standard, the visual terminal will display corresponding prompts, such as swinging the arms forward 3cm, swinging 15 degrees to the right, etc.

[0130] When a dance move scores low, the visual terminal will ask whether to perform the dance move again.

[0131] The specific reminder information reminder method and content are not limited and can be preset in advance. It is a well-known technology to issue a reminder instruction to another independent unit based on a certain judgment result. The specific judgment steps are not described in detail in this solution.

[0132] The present solution also provides an interactive online dance teaching device, comprising a display screen 1 and a dance mat 2. The display screen 1 is provided with a speaker and has a built-in memory and a controller. The memory stores dance video information of a plurality of preset dances. The controller has a built-in program that can run the interactive dance teaching method in claims 1-9. The dance mat 2 is arranged at the front end of the display screen 1, and has a plurality of pressure sensors regularly distributed therein that are connected to the controller for communication. In the present solution, 2x2 pressure sensors or pressure sensors with the same number of length and width and equidistant distribution can be set in each cell.

[0133] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. 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 will not cause the essence of the corresponding technical solutions to deviate from the protection scope of the technical solutions of the various embodiments of the present invention.

Claims

1. An interactive online dance teaching method, characterized in that: Based on a visual terminal capable of running an interactive dance teaching system, the interactive dance teaching system includes a collection module and a reminder module; The acquisition module includes an image acquisition unit, a tactile feedback unit and a processing unit; The image acquisition unit is used to acquire dynamic images of the user's posture in real time; The tactile feedback unit determines the user's standing position based on the user's real-time standing position, and is located on the ground and in front of the image acquisition unit. It includes a 3x3 matrix grid, and each matrix grid is provided with regularly arranged pressure sensing sub-units; The processing unit determines the user's real-time dance posture information based on the collected dynamic posture image and standing position, and sends the real-time dance posture information to the visualization terminal. The visualization terminal creates a dynamic virtual user image based on the user's real-time dance posture information and compares the real-time dance posture information with preset dance video information. The dance video information is displayed in the form of a virtual dancer's dance posture. The visualization terminal generates a reminder signal based on the comparison result between the real-time dance posture information and the preset dance video information. The reminder module sends an instruction to the user according to the reminder signal. The comparison result includes the user's incorrect posture interval information and incorrect standing position information; the steps for creating the dynamic virtual user image are as follows: S1: First, collect the user's dance video within a certain time interval and identify the user's body information in each frame of the video; S2: Secondly, the user's foot posture and standing position are determined by the number and distribution of the pressure sensing sub-units triggered in the tactile feedback unit; S3: The visualization terminal extracts the user's limb information in step S1 and the user's foot posture and standing position in step S2 to generate a virtual user image, and accelerates multiple consecutive frames of the virtual user image to form a video; The user's limb information includes the positions of the user's hands and feet; the steps of comparing the real-time dance posture information with the preset dance video information are as follows: T1: Create a three-dimensional coordinate system with the center of the matrix in the tactile feedback unit as the origin, the length and width of the matrix as the x-axis and y-axis respectively, and the user's height as the z-axis. Mark the user's limb information in step S1 in the three-dimensional coordinate system. T2: Divide the preset dance video information into dance pictures frame by frame according to the time interval in step S1, and extract the key points of the limb positions in the dance pictures and mark them in the three-dimensional coordinate system in step T1; T3: Compare the marking points in steps T1 and T2. If there is a deviation between the user's limb information and orientation marks and the limb position key points in the preset video information, it is considered as a non-standard dance posture and counted; The preset dance video information is any frame image in the dance video reserved in the visualization terminal, and the user's limb information and limb position key points extracted are all captured from the same time node and are all located in the same complete dance movement; the calculation method of the incorrect posture interval information is as follows: T301: Let any coordinate point in the user's limb information in step T1 be m, and the coordinates of m are (x0, y0, z0). Let the part of the limb position key point corresponding to m in step T2 be n, and the coordinates of n are (x1, y1, z1). T302: Set the deviation threshold length to l; T303: When and If all the above conditions are met, it is regarded as a standard posture. Otherwise, it is regarded as a non-standard posture and counted, which is recorded as p (p is an integer greater than 0). The visualization terminal also includes a counting unit for determining whether any complete dance movement is standard. The specific determination steps are as follows: Step 1: Based on the same complete dance action, extract video clips from the real-time dance posture information and dance video information respectively; Step 2: Select several image segments at the same time point from the real-time dance posture information video segment and the dance video information video segment obtained in step 1; Step 3: Set the number of non-standard postures allowed in a complete dance move to q (q is an integer greater than 0); Step 4: Compare the number of times p appears in the image segments obtained in step 2 after calculating the wrong posture interval information with q; When p≤q, the complete dance movement is considered to meet the standard; otherwise, the complete dance movement is considered to not meet the standard; and after step T303, the following steps are further included: T304: Set the gradient comparison threshold, respectively and l; T304: Compare the user's limb information in step T303 with the deviation values ​​of the key position points of the limbs, and score them respectively; when 3 points; when 3 points; when 3 points, when 2 points; when 2 points; when 2 points; when 1 point; when 1 point; when 1 point; when and If the value of is greater than 1, no points will be scored; T305: Count the scores of each complete dance move in the entire dance video information.

2. The interactive online dance teaching method according to claim 1, characterized in that: The visualization terminal can display a dynamic virtual user image established based on the user's real-time dance posture information and a virtual dancer image on the same screen, or blur the virtual dancer image and display it with the dynamic virtual user image as a double image.

3. An interactive online dance teaching method according to claim 2, characterized in that: The visualization terminal can adjust the size of the virtual dancer's image in proportion to the user's dynamic virtual user image.

4. The interactive online dance teaching method according to claim 1, characterized in that: The indication information includes at least one or more of voice reminder, visual terminal reminder, and light conversion.

5. An interactive online dance teaching device, characterized in that: include: A display screen (1), wherein the display screen (1) is provided with a speaker and has a built-in memory and a controller, wherein the memory stores dance video information of a plurality of preset dances, and the controller has a built-in program capable of running the interactive dance teaching method described in claims 1 to 4; A dance mat (2) is provided at the front end of a display screen (1), and has a plurality of pressure sensors regularly distributed therein and connected to a controller for communication.

Images