A physical fitness test method, device, computer equipment and storage medium

Abstract

The present disclosure provides a physical fitness test method, apparatus, computer equipment and storage medium, wherein the method comprises: obtaining motion video data of a target human body performing a target motion corresponding to the physical fitness test during the physical fitness test; performing key point detection on the motion video data to obtain key point detection results corresponding to the motion video data; and generating test result data of the target human body performing the physical fitness test based on the key point detection results, wherein the test result data is used to reflect the physical fitness condition of the target human body in at least one measurement dimension.

Description

Technical Field

[0001] The present disclosure relates to the field of computer vision technology, and in particular to a physical fitness testing method, apparatus, computer equipment and storage medium. Background Art

[0002] Students' physical fitness has received a lot of attention in all stages of education. Holding a series of sports and fitness activities, sports competitions, and sports tests can allow students to intuitively understand their own physical fitness. However, most physical fitness conditions are usually determined by coaches, physical education teachers, etc. through manual judgment. The manual determination of test results often leads to inaccurate test results due to factors such as different measurement standards and differences in subjective evaluations. Summary of the invention

[0003] The embodiments of the present disclosure at least provide a physical fitness testing method, apparatus, computer equipment and storage medium.

[0004] In a first aspect, an embodiment of the present disclosure provides a physical fitness testing method, comprising:

[0005] Acquiring motion video data of a target human body performing a target motion corresponding to the physical fitness test during a physical fitness test;

[0006] Performing key point detection on the motion video data to obtain key point detection results corresponding to the motion video data;

[0007] Based on the key point detection results, test result data of the target person when undergoing a physical fitness test is generated, and the test result data is used to reflect the physical fitness condition of the target person in at least one measurement dimension.

[0008] In this way, machine vision technology is used to accurately obtain key point detection results of motion image data obtained from the target human body during the physical fitness test, and the key point detection results are used to generate test result data with higher accuracy.

[0009] In a second aspect, an embodiment of the present disclosure further provides a physical fitness testing device, comprising:

[0010] An acquisition module, used to acquire motion video data of a target human body performing a target motion corresponding to the physical fitness test during a physical fitness test;

[0011] A detection module, used to perform key point detection on the motion video data to obtain key point detection results corresponding to the motion video data;

[0012] A generation module is used to generate test result data of the target human body when performing a physical fitness test based on the key point detection results; the test result data is used to reflect the physical fitness condition of the target human body in at least one measurement dimension.

[0013] In a third aspect, an optional implementation of the present disclosure further provides a computer device, a processor, and a memory, wherein the memory stores machine-readable instructions executable by the processor, and the processor is used to execute the machine-readable instructions stored in the memory, and when the machine-readable instructions are executed by the processor, the machine-readable instructions perform the steps in the above-mentioned first aspect, or any possible implementation of the first aspect.

[0014] In a fourth aspect, an optional implementation of the present disclosure further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed, the steps of the above-mentioned first aspect, or any possible implementation of the first aspect are executed.

[0015] For a description of the effects of the above-mentioned apparatus, computer equipment and computer-readable storage medium, please refer to the description of the above-mentioned physical fitness testing method, which will not be repeated here.

[0016] It should be understood that the above general description and the following detailed description are merely exemplary and explanatory, and are not intended to limit the technical solutions of the present disclosure.

[0017] In order to make the above-mentioned objectives, features and advantages of the present disclosure more obvious and easy to understand, preferred embodiments are specifically cited below and described in detail with reference to the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the following is a brief introduction to the drawings required for use in the embodiments. The drawings herein are incorporated into the specification and constitute a part of the specification. These drawings illustrate embodiments consistent with the present disclosure and are used together with the specification to illustrate the technical solutions of the present disclosure. It should be understood that the following drawings only illustrate certain embodiments of the present disclosure and should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can also be obtained based on these drawings without creative work.

[0019] Figure 1 A flow chart of a physical fitness testing method provided by an embodiment of the present disclosure is shown;

[0020] Figure 2 A specific example of a key point of a human body provided by an embodiment of the present disclosure is shown;

[0021] Figure 3A specific example of the relative positions between key points of a target human body when performing yoga exercises provided by an embodiment of the present disclosure is shown;

[0022] Figure 4 A schematic diagram of a physical fitness testing device provided by an example of the present disclosure is shown;

[0023] Figure 5 A schematic diagram of a computer device provided by an example of the present disclosure is shown. DETAILED DESCRIPTION

[0024] In order to make the purpose, technical scheme and advantages of the embodiments of the present disclosure clearer, the technical scheme in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are only part of the embodiments of the present disclosure, rather than all of the embodiments. The components of the embodiments of the present disclosure generally described and shown here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present disclosure is not intended to limit the scope of the present disclosure claimed for protection, but merely represents the selected embodiments of the present disclosure. Based on the embodiments of the present disclosure, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present disclosure.

[0025] Research has found that in school physical test activities, due to the lack of professional testing equipment, the results of physical tests are usually determined by the manual judgment of coaches or physical education teachers. For example, in the strength test, male students are required to do push-ups or pull-ups within a certain period of time, and the number of push-ups or pull-ups is used to determine whether their strength reaches a certain standard; while female students are required to do sit-ups within a certain period of time, and the number of sit-ups is used to determine whether their strength reaches a certain standard. For example, in the sensitivity test, students are required to run around the pole, run with a ball in a curve, and jump in the cross quadrant to determine the students' sensitivity information. In the test process, the manual judgment method usually cannot have a unified measurement standard, and the subjective evaluation has certain differences, resulting in inaccurate test results.

[0026] Based on the above research, the present disclosure provides a physical fitness testing method. Without the need for professional testing equipment, machine vision technology is used to accurately obtain the key point detection results of the motion video data obtained by the target human body during the physical fitness test. Under the same measurement standard, the key point detection results are used to generate test result data, thereby reducing the differences caused by subjective factors and having higher accuracy.

[0027] The defects of the existing solutions and the solutions proposed in the present disclosure are the results obtained by the inventor after practice and careful research. Therefore, the discovery process of the above problems and the solutions proposed by the present disclosure in this article should be the contributions made by the inventor to the present disclosure during the disclosure process.

[0028] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.

[0029] To facilitate understanding of the present embodiment, a physical fitness test method disclosed in the embodiment of the present disclosure is first introduced in detail. The execution subject of the test method provided in the embodiment of the present disclosure is generally a computer device with certain computing capabilities, and the computer device includes, for example: a terminal device or a server or other processing device, and the terminal device can be a user equipment (User Equipment, UE), a mobile device, a user terminal, a terminal, a cellular phone, a cordless phone, a personal digital assistant (Personal Digital Assistant, PDA), a handheld device, a computing device, a vehicle-mounted device, a wearable device, etc. In some possible implementations, the test method can be implemented by a processor calling a computer-readable instruction stored in a memory.

[0030] The test method provided in the embodiment of the present disclosure can be applied to a variety of physical fitness test scenarios, such as fitness scenarios, competition scenarios, school sports test scenarios, etc. The method can be used to perform physical fitness tests on group users or on individual users. The specific scenario of use can be determined according to actual needs, and the embodiment of the present disclosure is not limited. In addition, the test method provided in the embodiment of the present disclosure can also be applied to a variety of sports and fitness exercise types, such as running, yoga, push-ups, sit-ups, pull-ups, running around a pole, etc., which are not limited in the embodiment of the present disclosure.

[0031] The testing method provided by the embodiment of the present disclosure is described below.

[0032] See also Figure 1 FIG. 1 is a flowchart of a testing method provided by an embodiment of the present disclosure, wherein the method includes steps S101 to S103, wherein:

[0033] S101, obtaining motion video data of a target human body performing a target motion corresponding to the physical fitness test during a physical fitness test.

[0034] S102: Perform key point detection on the motion video data to obtain key point detection results corresponding to the motion video data.

[0035] S103. Based on the key point detection results, generate test result data of the target person when undergoing a physical fitness test, wherein the test result data is used to reflect the physical fitness condition of the target person in at least one measurement dimension.

[0036] In the disclosed example, after obtaining motion video data of the target human body performing the target motion corresponding to the physical fitness test during the physical fitness test, key point detection is performed on the image data to obtain the key point detection results corresponding to the image data, and based on the key point detection results, test result data of the target human body during the physical fitness test is generated, so that when the target human body performs the target motion, the target human body key points related to the target motion are detected, and the corresponding test result data is generated based on the relevant human body key point detection results, and then the key point detection results of the motion video data obtained by the target human body during the physical fitness test are accurately obtained by using machine vision technology, and the key point detection results are used to generate test result data with higher accuracy.

[0037] The above S101 to S103 are described in detail below.

[0038] Regarding the above S101: The testing method in the disclosed example can be applied to different physical fitness tests, and the physical fitness test can include at least one of the following: speed test, endurance test, strength test, sensitivity test, and flexibility test. When a user is taking a physical fitness test, the user taking the physical fitness test can be filmed by a terminal device such as a video recorder, a camera, or a mobile phone to obtain motion video data. In this process, the human body of the user taking the physical fitness test contained in the image data is the target human body.

[0039] The motion video data may include: video data and / or image data. In the case where the motion video data includes video data, the video data includes a plurality of frames of motion images.

[0040] When the test method provided by the embodiment of the present disclosure is executed in a terminal device, the terminal device can directly shoot the target human body to obtain motion video data; when the above test method is executed in a server, the server can receive the motion video data shot by the terminal device.

[0041] In the disclosed example, the type of target exercise corresponding to the physical fitness test can be predetermined.

[0042] Exemplarily, if the current physical fitness test is a speed test, the target exercise type can be set to at least one of the following exercise types: sprinting (such as 50-meter run, 100-meter run), race walking, etc.

[0043] If the current physical fitness test is an endurance test, the target exercise type can be set to at least one of the following exercise types: long-distance running, rope skipping, plank support, etc.

[0044] If the current physical fitness test is a strength test, the target exercise type can be set to at least one of the following exercise types: pull-ups, push-ups, standing long jumps, 15-minute supine leg raises, etc.

[0045] If the current physical fitness test is a sensitivity test, the target movement type can be set to at least one of the following movement types: curve dribbling, running around the pole, cross quadrant jump, etc.

[0046] If the current physical fitness test is a flexibility test, the target exercise type can be set to at least one of the following exercise types: high leg raises, yoga, etc.

[0047] In this way, without the need for professional testing equipment, machine vision technology can be used to accurately obtain the key point detection results of the motion video data obtained by the target human body during the physical fitness test. Under the same measurement standard, the key point detection results are used to generate test result data, reducing the differences caused by subjective factors and having higher accuracy.

[0048] Regarding the above S102: In the example of the present disclosure, the key point detection result, for example, includes: position information corresponding to multiple human key points in the target human body.

[0049] When performing key point detection on the motion video data, for example, a human key point detection model may be used to implement the detection.

[0050] The human key point detection model includes, for example, at least one of the following: a limb key point detection model, a head and shoulder key point detection model, a hand frame point model, etc. For different types of target motion, the key point detection models used can be the same or different, depending on the key point positions and / or key point quantities required for the actual motion type.

[0051] like Figure 2 As shown in the figure, a specific example of 18 human body key points of the target human body is obtained after detecting the motion video data of the target human body through the limb key point model.

[0052] Among them, the head and shoulders include four key points of the human body, which correspond to Figure 2 The numbers in the figure are 0, 1, 2, and 3; the hand frame includes four key points of the human body, which correspond to Figure 2 The numbers in are 4, 5, 6, and 7.

[0053] The motion video data, for example, includes: multiple frames of motion images. The key point detection results include: key point detection results corresponding to the multiple frames of motion images.

[0054] In this case, key point detection may be performed on multiple frames of motion images respectively to obtain key point detection results corresponding to each frame of motion images.

[0055] Regarding S103: In a specific implementation, when generating test result data of a target person during a physical fitness test based on the key point detection results, for example, the following method may be used:

[0056] Based on the movement type of the target movement, a target human body key point matching the target movement is determined from the multiple human body key points of the target human body.

[0057] Based on the key point detection result, position information corresponding to the key points of the target human body is determined.

[0058] Based on the position information corresponding to the key points of the target human body, test result data of the target human body when performing physical fitness is generated.

[0059] Among them, for different types of sports, the movements are different. For example, if the target sport type is push-ups, the sport feature is to keep the shoulder and neck moving up and down. Therefore, the test result data of the target human body when doing push-ups can be obtained through the key points of the human body corresponding to the human shoulder.

[0060] For another example, if the target exercise is sit-ups, the movement characteristic of which is that the upper body is constantly standing up and alternating between lying down. Therefore, the test result data of the target human body when performing sit-ups can be obtained through the key points of the human body corresponding to the human torso.

[0061] Therefore, different target movements correspond to different target human body key points, and according to the specific movement type of the target movement, the target human body key points matching the target movement can be determined from the multiple human body key points of the target human body.

[0062] After determining the target human key points, the position information corresponding to the target human key points can be determined based on the key point detection results. The position information corresponding to the target human key points includes, for example: two-dimensional position information of the target human key points in the image coordinate system corresponding to the motion video data, and / or three-dimensional position information of the target human key points in the camera coordinate system corresponding to the motion video data.

[0063] The test result data of the target human body during the physical fitness test includes at least one of the following: the test result data corresponding to the speed test, the test result data corresponding to the endurance test, the test result data corresponding to the strength test, the test result data corresponding to the sensitivity test, and the test result data corresponding to the flexibility test.

[0064] A: For physical fitness tests including speed tests, the test result data includes the test result data corresponding to the speed test of the target human body:

[0065] The test result data corresponding to the speed test includes: speed information of a target human body when performing a running exercise within a target test range determined based on a running test distance, and the target exercise may include but is not limited to running exercise.

[0066] The speed information includes at least one of the following: an average moving speed, a maximum moving speed, a minimum moving speed, a first moving speed at a target movement moment, and a second moving speed at a target position when running within the target test range determined based on the running test distance.

[0067] Among them, the target movement moment can include any moment when the target person is running during the speed test; the target position can be any position within the target test range when the target person is running during the speed test.

[0068] a1: For the case where the speed information includes: the average moving speed of the target when the target moves within the target test range:

[0069] The target motion image includes: a first motion image in which the target human body is located at a starting position corresponding to the target test range, and a second motion image in which the target human body is located at an ending position corresponding to the target test range.

[0070] Among them, the first motion image refers to the motion image corresponding to the starting position corresponding to the target test range; the second motion image refers to the motion image corresponding to the end position corresponding to the target test range; the first motion image and the second motion image are used to determine the time taken for the target human body to move from the starting position to the end position, that is, to pass through the target test range.

[0071] In a specific implementation, the first motion image and the second motion image can be obtained by one of the following methods:

[0072] Example 1: The camera is fixedly installed so that the camera's field of view can just cover the target test range from a side perspective, that is, one end of the camera's shooting field of view is the starting position, and the other end is the end position; at this time, if the starting point of the target test range is on the left, the motion image when the target human body is located at the leftmost pixel point of the motion video data can be used as the first motion image, and the motion image when the target human body is located at the rightmost pixel point of the motion video data can be used as the second motion image; if the starting point of the target test range is on the right, the opposite is true.

[0073] Example 2: The camera is fixedly installed so that the target test range is within the field of view of the camera, and markers, such as horizontal lines on the ground or other markers, can be set at the starting position and the end position corresponding to the target test range; based on the key point detection results corresponding to each frame of the target motion image, the motion image of the target human body at the position of the marker corresponding to the starting position indicated by the key point detection result is used as the first motion image, and the motion image of the target human body at the position of the marker corresponding to the end position indicated by the key point detection result is used as the second motion image.

[0074] Among them, the method for determining the target human body position in the target motion image can be through the following method: based on the key point detection result, the position information of all human body key points is obtained, and then the position of the human body center point is determined according to the position information of all human body key points.

[0075] Alternatively, the positions of certain specific key points of the human body among the multiple key points are used as the positions of the target human body. The specific key points include, for example, key points of the human body trunk position (such as Figure 2 The key points with the label number 8), the key points closest to the start / end point, the hand key points (such as Figure 2 The key points numbered 6 and 7), the key points of the legs (such as Figure 2 At least one of the key points numbered 12 and 13).

[0076] In response to determining the first motion image and the second motion image, the time when the target human body passes through the target test range is obtained based on the frame image timestamps corresponding to the first motion image and the second motion image, and the target test range and the time when the target human body passes through the target test range are used to obtain the average moving speed of the target human body when performing the target motion within the target test range.

[0077] In addition, taking the target sport of 100-meter running as an example, the target test range described in the embodiment of the present disclosure can be 100 meters, or it can be a partial distance in 100 meters, for example, the distance of 0m-10m in the 100m test is used as the target test range, or the distance of 15m-35m is used as the target test range.

[0078] In a specific implementation, the moving distance corresponding to each frame of the motion image can be determined by the key point detection results corresponding to each frame of the motion image, and the key point detection results corresponding to the adjacent motion images that are adjacent to each frame of the motion image in front and / or behind the timestamp, and the moving distance corresponding to each frame of the motion image can be determined based on the proportional relationship between the predetermined image distance and the actual physical distance, and the displacement of the target human body in the image coordinate system corresponding to the image data. The image size of the target motion image is determined, and after the pixel positions of the starting point and the end point in the target motion image are determined by the above method, the position of the target test range in the target motion image is determined based on the total number of pixels in the target test range and the pixels within a partial distance in the target test range. For example: in a 100m sprint test, the number of horizontal pixel points in the target motion image is 4000, the 1000th horizontal pixel point is the starting point, and the 3000th horizontal pixel point is the end point; at this time, the test distance from the 1000th horizontal pixel point to the 3000th horizontal pixel point is 0m-100m; correspondingly, the 2000th horizontal pixel point is 50m; based on this method, the horizontal pixel points corresponding to the starting position and the end position of any test distance in the target motion image can be determined.

[0079] Among them, since the target key point and the starting point / end point may correspond to multiple pixel points in the motion image, the leftmost or rightmost pixel point among the corresponding multiple horizontal pixel points can be uniformly taken as the pixel point for judgment, and the proportional relationship between the image distance and the actual physical distance can be determined according to the shooting device used, for example, it can be the field of view (angle of view) parameter of the camera. In optical instruments, the angle formed by the two edges of the maximum range of the object image of the measured target that can pass through the lens with the lens of the optical instrument as the vertex is called the field of view angle. The size of the field of view angle determines the field of view range of the optical instrument. The larger the field of view angle, the larger the field of view and the smaller the optical magnification. In layman's terms, the target object will not be collected in the lens if it exceeds this angle.

[0080] In this way, the time information and distance information of the target human body during exercise can be obtained through the relationship between the image distance and the actual distance, as well as the time information corresponding to each frame of the motion image, thereby obtaining the speed information of the target human body. In at least one dimension of the average moving speed during running, the maximum moving speed, the minimum moving speed, the first moving speed at the target movement moment, and the second moving speed at the target position, the physical fitness of the target human body in terms of speed can be more comprehensively reflected.

[0081] In an optional implementation, if the width (W) of the imaging plane is fixed, the size of the field of view angle is directly determined by the focal length. The larger the focal length, the farther you can see, but the smaller the field of view angle. The smaller the focal length, the closer you can see, but the larger the field of view angle.

[0082] For example, when the total horizontal pixel number of the image is 4000, if the field of view angle is 80 degrees, the starting point is located at the 1000th horizontal pixel point, and the field of view angle is 103 degrees. Since the total resolution of the image remains unchanged, the number of horizontal pixels will not change, and the field of view angle increases, that is, the field of view increases, so the objects in the camera screen will appear farther away from the person, that is, the starting point can be changed to the 1500th horizontal pixel point.

[0083] For example, in a 200m test, when a camera with 4000 horizontal pixels of the obtained motion image is used for shooting, the 1000th horizontal pixel can be obtained as the starting point based on the target detection algorithm, and the 3000th horizontal pixel can be obtained as the end point. The number of pixels corresponding to the actual distance covered by the camera in the captured image will change according to the different field of view angles. When implementing this method, the corresponding relationship between the actual distance and the image distance needs to be flexibly transformed according to the different field of view angles.

[0084] When the target key point in a frame of multiple frames of motion images appears at or is closest to the 1000th horizontal pixel point where the starting point is, the frame of motion image is determined as the first motion image; when the target key point in a frame of multiple frames of motion images appears at or is closest to the 3000th horizontal pixel point where the end point is, the frame of motion image is determined as the second motion image.

[0085] Based on the time corresponding to the first motion image and the second motion image on the timestamp, for example: the time of the first motion image on the timestamp is 00:20, and the time of the second motion image on the timestamp is 00:40, then it can be obtained that the average speed of the target human body at a test distance of 200 meters is 10 meters per second.

[0086] If you need to get the average speed of other test distances, for example, the average speed of the target human body from 90 meters to 110 meters, you can assume that the horizontal pixel points corresponding to 0 meters and 200 meters are the 1000th and 3000th, and the horizontal pixel points corresponding to 90 meters and 110 meters are the 1900th and 2100th.

[0087] In response to the current target horizontal pixel points being the 1900th and 2100th, the frame image when the target key point appears at the 1900th pixel point or the frame image closest to the pixel point is determined as the current first motion image, and the frame image when the target key point appears at the 2100th pixel point or the frame image closest to the pixel point is determined as the current second motion image. The average time calculation method within the total target test range provided by the embodiment of the present disclosure can also obtain the average time taken by the target human body to pass the distance of 90 meters to 110 meters, which will not be repeated here.

[0088] a2: In the case where the speed information includes the first moving speed of the target human body at the target movement moment during the target movement, the target motion image includes: the third motion image whose timestamp in the motion video data is closest to the target movement moment, and the fourth motion image adjacent to the third motion image in timestamp.

[0089] The first moving speed of the target at the moment of movement refers to the instantaneous speed at a certain moment.

[0090] The third motion image refers to a motion image at a specific moment in the motion video data when the target human body performs the target motion, wherein, since the video frames per second obtained by different parameter camera parameter settings are different, a frame of motion image with a timestamp closest to the target motion moment is taken as the third motion image. The third motion image is used to determine the first moving speed corresponding to the target motion moment that needs to be obtained.

[0091] The fourth motion image includes the two frames of images adjacent to the third motion image. For example, at the 5th second when the target human body performs the target motion, the camera parameters set the motion video data to shoot 60 frames of images per second. Then, the 30th frame of the 60 frames of motion images shot at the 5th second can be taken as the third motion image, and the fourth motion image is the 29th and 31st frames.

[0092] In a specific implementation, in response to determining the target moment at which the first moving speed needs to be obtained, based on the method for determining the position of the horizontal pixel points where the human target key points are located in the average speed provided in the embodiment of the present disclosure, the nth horizontal pixel point of the human target key points in the motion image at the target moment is obtained, and the frame image is determined as the third motion image, and the two frames before and after the frame image are determined as the fourth motion image.

[0093] Among them, since the target key point may occupy multiple horizontal pixel points in the motion image, the leftmost or rightmost pixel point among the corresponding multiple horizontal pixel points can be uniformly taken as the pixel point used for judgment.

[0094] Based on the horizontal pixel positions of the target key points in the third motion image and the fourth motion image, the horizontal pixel displacement difference between the third motion image and the two frames of images before and after is obtained; based on the pixel displacement difference, the moving distance corresponding to the pixel displacement difference can be obtained; based on the timestamps corresponding to the third motion image and the fourth motion image, the time difference is obtained; based on the moving distance and time difference between the third motion image and the fourth motion image, the first moving speed at the target motion moment when the target human body performs the target motion within the target test range is determined.

[0095] a3: In response to the speed information including: a second moving speed of the target position when the target motion is performed within the target test range, the target motion image includes: a fifth motion image in which the position of the target human body in the motion image is closest to the target position, and a sixth motion image adjacent to the fifth motion image in terms of timestamp.

[0096] Among them, the fifth motion image refers to a motion image of a specific position in the motion video data of the target human body performing the target motion, and in the motion image, the specific position corresponds to a specific pixel point; regarding the method of determining the horizontal pixel points at any position through the horizontal pixel points at the starting point and the end point in the motion image, it has been described in detail in the method for determining the average speed provided in the embodiment of the present disclosure, and will not be repeated here.

[0097] The sixth moving image includes two frame images preceding and following the fifth moving image.

[0098] In a specific implementation, in response to determining the target position that needs to obtain the second moving speed, the horizontal pixel points in the target motion image corresponding to the target position are determined based on the method for determining the positions of the horizontal pixel points where the key points of the human target are located in the average speed provided in the embodiment of the present disclosure.

[0099] A frame image in which the position of the target key point is closest to the target position is selected from the multiple frames of motion images as the fifth motion image.

[0100] The two frames of images before and after the fifth motion image are used as the sixth motion image.

[0101] Based on the horizontal pixel positions of the target key points in the fifth motion image and the sixth motion image, the horizontal pixel displacement difference between the fifth motion image and the two frames of images before and after is obtained; based on the pixel displacement difference, the moving distance corresponding to the pixel displacement difference can be obtained; based on the timestamps corresponding to the fifth motion image and the sixth motion image, the time difference is obtained; based on the moving distance and time difference corresponding to the fifth motion image and the sixth motion image, the second moving speed at the target mark position when the target human body performs the target motion within the target test range is determined.

[0102] a4: For the case where the speed information includes the maximum moving speed and the minimum moving speed of the target human body when the target moves:

[0103] Based on the method for obtaining the first movement speed and the second movement speed provided in the embodiment of the present disclosure, the movement speeds of the target human body corresponding to the multiple frames of motion images when the target human body is moving can be obtained. Then, based on the movement speeds corresponding to the multiple frames of images, the maximum speed among them is determined as the maximum movement speed of the target human body when the target human body is moving, and the minimum speed among them is determined as the maximum movement speed of the target human body when the target human body is moving.

[0104] In a possible implementation, in addition to the above speed, two frames of images may be randomly extracted from the image data in the above manner to determine the speed.

[0105] For example, if the video frame rate is 30 frames per second and the total motion time of the target human body is 10 seconds, the total video frame rate is 300 frames. At this time, any two frames of images can be extracted from the 300 frames to calculate the corresponding speed, such as the 31st frame and the 90th frame. According to the actual distance calculation method mentioned above, the actual motion distance of the target user in the two frames of images is obtained by the number of horizontal pixels in the obtained images, as well as the time information corresponding to the two frames of images, to obtain the required speed information, such as the average speed information and instantaneous speed information at different time points or time periods. As long as the motion time information and motion distance information of the target human body are confirmed, the above speed information can be obtained by calculation.

[0106] B: The physical fitness test includes an endurance test, and the test result data includes the test result data corresponding to the endurance test of the target human body. The target exercise includes at least one of the following: long-distance running, rope skipping, and plank support.

[0107] In order to obtain the test result data corresponding to the endurance test, the corresponding target human body posture information can be obtained from the target human body in multiple movement stages of the target movement; the target human body posture information in the corresponding stage is compared with the posture information of the target human body at the beginning of the target movement, and finally the test result data corresponding to the corresponding endurance test is obtained.

[0108] In this way, the first postures corresponding to the multiple movement stages and the standard posture information corresponding to the target movement can be used to more accurately reflect the physical fitness of the target human body in terms of endurance.

[0109] In a specific implementation, when determining the endurance data of the target human body when performing the target exercise, the following steps 11 to 12 may be performed:

[0110] Step 11: Based on the key point detection result, determine the first posture information corresponding to the multiple movement stages of the target movement when the target human body performs the target movement.

[0111] The first posture information refers to, for example, actual posture information of the target human body when exercising, and / or actual posture information of a target limb of the target human body.

[0112] The plurality of movement stages are determined based on the duration of the target movement or based on the moving distance of the target movement; for example, in an endurance test of a 4000m long-distance run, 0-1000, 1000-2000, and 2000-4000 can be divided into three stages, or the time information of the target human body performing the target movement can be used for division; here, the division of different movement stages based on the moving distance of the target movement is taken as an example:

[0113] For different exercise stages, assuming that it is necessary to obtain the endurance data corresponding to the target human body at 1000m, 2000m, and 4000m respectively, then based on the key point detection results of the target motion image corresponding to the specific exercise stage, the first posture information of the target human body at 1000m, 2000m, and 4000m is obtained.

[0114] Step 12: Based on the first posture information corresponding to the plurality of movement stages respectively and the standard posture information corresponding to the target movement, determine the endurance data of the target human body when performing the target movement.

[0115] Among them, the standard posture information refers to the standard posture when performing the target movement. This standard posture can be entered based on different evaluation criteria, or it can be the starting posture of the target body when performing the target movement. For example, in long-distance running, a specific value of the swing amplitude of both arms can be set as the standard posture information.

[0116] For the multiple first posture information of the target human body corresponding to different motion stages obtained in step one, the first posture information corresponding to the different motion stages is compared with the standard posture information to obtain similarity information between the first posture information and the standard posture information. For example, in long-distance running, as the long-distance running progresses, the endurance of the target human body decreases, the swing amplitude of both arms decreases accordingly, and the similarity between the determined first posture information and the standard posture information also decreases accordingly; therefore, the swing amplitude of both arms in the first posture information corresponding to different motion stages can be compared with the standard swing amplitude of the standard posture information to determine the similarities corresponding to the target human body when performing the target motion in multiple motion stages; the higher the similarity, the higher the endurance; the lower the similarity, the lower the endurance.

[0117] Exemplarily, the method for testing the endurance data of the target human body when performing the target sport can be applied to a variety of target sports, such as long-distance running, rope skipping, plank support, etc.

[0118] Taking long-distance running as a target sport, in the 1600m long-distance running test, the standard posture is set to an arm swing amplitude of positive 45° (the swing amplitude when the arms are perpendicular to the ground is regarded as 0°). This standard posture setting does not take into account the amplitude of the arms when swinging backwards, because the forward and backward swing amplitudes of most people's arms are not consistent when running, so only the forward swing is considered.

[0119] For example, when a target person runs a 400m lap in a standard track and field venue, endurance data at 400m, 800m, and 1600m need to be obtained. At this time, the camera can be placed at the starting point, and the key points of the target person appearing in the camera's field of view are detected. The first time the target person appears in the camera's field of view is 0m, the second time the target person appears in the camera's field of view is 400m, the third time the target person appears in the camera's field of view is 800m, and the fourth time the target person appears in the camera's field of view is 1600m (considering that the movements when crossing the finish line in actual long-distance running may vary too much, the camera position can be appropriately moved to reduce the error).

[0120] When the target human body appears in the camera's field of view for the second time, the target key point detection result corresponding to the motion stage is determined, and the first posture information corresponding to the motion stage is determined. For example, when the target human body appears in the camera's field of view for the second time, the target human body exists in multiple frames of motion images in the motion video data, and the frame image with the maximum amplitude of the arm swing can be taken to determine the first posture information.

[0121] Based on the key point detection results, the opening and closing angles of the human torso and the upper side of the arms in the frame image with the maximum amplitude of the arm swing are determined; taking 30° as an example, the arm swing angle in the first posture information is 30°; through the arm swing angle of 30° in the first posture information and the arm swing angle of 45° in the standard posture information, the posture standard degree information obtained is approximately 66%.

[0122] Based on the obtained posture standard degree information, the endurance data of the target human body when performing the target movement is determined. For example, if the endurance value is 100, the endurance value of the target human body at 400m can be 66. In the specific implementation, different weights can also be added to affect the endurance data finally presented. The methods at 800m and 1600m are the same and will not be repeated here.

[0123] In addition, taking plank support as a target exercise as an example, in the plank support exercise, multiple exercise stages of the target exercise can be determined based on the duration of the target exercise. For example, the exercise stages can be divided into 1 minute, 5 minutes, and 10 minutes.

[0124] In a specific implementation, the camera can be placed at the side of the human body where the entire human body can be photographed, and the target human body doing the plank support can be photographed at the 1st minute, the 5th minute, and the 10th minute respectively.

[0125] The key points of the human body are detected for multiple frames of motion images in multiple motion stages respectively; the first posture information is determined; the angle between the upper arm and the forearm is taken as a specific angle as the standard posture, and the posture standard degree information is obtained by comparing the angle between the upper arm and the forearm in the first posture information with the angle between the upper arm and the forearm in the standard posture information; based on the obtained posture standard degree information, the endurance data of the target human body when performing the target motion is determined.

[0126] In addition, for each frame of motion image during the target motion, the first posture information corresponding to each frame of motion image can be determined, and the similarity between the first posture information corresponding to each frame of motion image and the standard posture information can be determined. Then, the endurance data can be obtained based on the similarities corresponding to all motion images.

[0127] The endurance value of the target human body represented by the endurance data is positively correlated with the target movement time, positively correlated with the movement speed of the target movement, and positively correlated with the similarity.

[0128] That is, assuming that the exercise time and exercise speed remain unchanged, the higher the similarity, the higher the endurance value.

[0129] When the similarity and movement time remain unchanged, the higher the movement speed, the higher the endurance value.

[0130] When the movement speed and similarity remain unchanged, the longer the exercise time, the higher the endurance value.

[0131] C: For the physical fitness test including a strength test, the test result data include the test result data corresponding to the strength test of the target human body when performing the target exercise: the target exercise includes at least one of the following: pull-ups, push-ups, standing long jump, and 10-minute supine leg raises.

[0132] In order to obtain the test result data corresponding to the strength test, the number of times the target human body performs the target movement can be determined through the key point detection results corresponding to the motion video data within a specific time.

[0133] Based on the number of times the target human body performs the target movement, strength data of the target human body when performing the target movement is determined; wherein the strength value of the target human body represented by the strength data is positively correlated with the number of times.

[0134] In a specific implementation, when determining the strength data of the target human body when performing the target movement, the following process may be performed:

[0135] In each counting cycle of at least one counting cycle, the following steps 21 to 24 are performed:

[0136] Step 21: In the current counting cycle, determine, from the multiple frames of images, a first reference image corresponding to the starting posture of the target human body when performing the target motion.

[0137] Step 22: Use other images whose timestamps are later than the first reference image as first alternative images, and determine a second reference image corresponding to the terminal posture when performing the target movement based on the key point detection results corresponding to the first reference image and the key point detection results corresponding to the alternative images.

[0138] Step 23: In response to determining that the second reference image is successful, updating the number of times the target motion is performed.

[0139] Step 24: Based on the number of times the target human body performs the target movement within a preset time period and in combination with a prescribed evaluation criterion, determine the strength data of the target human body when performing the target movement.

[0140] Among them, for the above step 21:

[0141] The motion video data includes multiple frame images, that is, a set composed of multiple frame images. In this set, the first reference image is first determined based on the key point detection results and the timestamp sequence of each frame image, and then other reference images required for determining the target motion count information are determined from the set based on the first reference image.

[0142] The first reference image refers to a frame of image in the motion video data corresponding to the starting posture of the target human body when performing the target motion.

[0143] In a specific implementation, for the first counting cycle: the first frame image in the motion video data can be directly used as the first reference image corresponding to the first counting cycle, and the posture of the target human body in the first frame image can be used as the starting posture corresponding to the target motion.

[0144] Example 1: If the motion image data includes: a1, a2, ..., an, a total of n frames of images, a1 can be directly used as the first reference image corresponding to the first counting cycle.

[0145] This solution is simple and easy to implement, requires less calculation, can reduce the computing power requirements of terminal devices, and improve processing efficiency. It is suitable for target sports with relatively simple movements, such as push-ups and pull-ups.

[0146] In addition, based on the target motion type, the starting posture corresponding to the target motion can be determined, and then the human body posture corresponding to each frame image can be determined based on the results of key point detection on the motion video data. Then, based on the human body posture corresponding to each frame image, a frame image with a human body posture consistent with the starting posture corresponding to the target motion and the earliest timestamp is determined from multiple frames of images as the first reference image for the first counting cycle.

[0147] Example 2: If the motion video data includes: a1, a2, ..., an, a total of n frames of images, then the human body posture corresponding to a1 can be determined based on the key point detection result corresponding to a1; if the human body posture corresponding to a1 is not the starting posture corresponding to the target motion, then the human body posture corresponding to a2 can be determined based on the key point detection result corresponding to a2; if the human body posture corresponding to a2 is not the starting posture corresponding to the target motion, then the human body posture corresponding to a3 can be determined based on the key point detection result corresponding to a3; if the human body posture corresponding to a3 is the starting posture corresponding to the target motion, then a3 is used as the first reference image corresponding to the first counting cycle.

[0148] This solution is more accurate in counting and is suitable for sports with more complex movements, such as dance with periodic movements.

[0149] For the above step 22: after determining the first reference image, other images whose timestamps are later than the first reference image of the first counting cycle can be used as first alternative images, and based on the key point detection results corresponding to the first reference image and the key point detection results corresponding to the alternative images, a second reference image corresponding to the terminal posture when performing the target movement is determined.

[0150] In the above example 1, in the first counting cycle, image a1 is used as the first reference image corresponding to the first counting cycle, and therefore a2~an are used as the first alternative images. Then, based on the key point detection results corresponding to a1 and the key point detection results corresponding to at least part of a2~an, the second reference image corresponding to the terminal posture during a target movement is determined.

[0151] The second reference image is, for example, a frame image corresponding to the end posture of the target human body when performing a target movement. The end posture can be pre-set according to the type of the target movement; for example, if the target movement is a push-up movement, the starting posture can be a state of supporting the ground with arms straight, and the end posture can be a state of supporting the ground with arms bent at a specific angle.

[0152] In this case, according to the human body posture corresponding to the first alternative image of each frame, the first alternative image whose human body posture is consistent with the terminal posture corresponding to the target motion and whose timestamp is the earliest frame among all the first alternative images that meet the conditions can be determined from multiple frames of first alternative images as the second reference image for the first counting cycle.

[0153] In addition, the target human body may also perform a target movement, and the posture when the body posture is restored to the starting posture is used as the ending posture, that is, the starting posture is the ending posture.

[0154] In this case, according to the human body posture corresponding to the first reference image of each frame, a frame image with the human body posture consistent with the terminal posture corresponding to the target movement and the earliest timestamp can be determined from multiple frames of first reference images as the second reference image of the first counting cycle. In addition, the positions of the key points of the target human body in the first reference image can be compared with the positions of the key points of the target human body in the first alternative image of each frame in turn; if the position difference between the key points of the target human body between the first reference image and a certain frame of the first alternative image is less than a preset threshold, and the timestamp of the first alternative image of the frame is the earliest timestamp among all images that meet the threshold condition, then the first alternative image is used as the second reference image.

[0155] The embodiment of the present disclosure also provides a specific method for determining a second reference image, including:

[0156] Step 221: Determine the current first candidate image from the first candidate images according to the order of timestamps.

[0157] Step 222: Based on the key point detection result corresponding to the first reference image and the key point detection result of the current first candidate image, determine whether the displacement of the target human key point corresponding to the target motion is greater than or equal to a preset second displacement threshold.

[0158] Step 223: In response to the displacement being greater than or equal to a preset second displacement threshold, determining the current first candidate image as the second reference image.

[0159] Step 224: In response to the displacement being less than the second displacement threshold, returning to the step of determining the current first candidate image from the first candidate images in the order of timestamps.

[0160] The second displacement threshold is the minimum value of the difference between the position coordinates of the multiple human body key points corresponding to the current posture of the target action and the position coordinates of the multiple human body key points corresponding to the terminal posture of the target action. When the difference between the position coordinates of the multiple human body key points corresponding to the current posture of the target action and the position coordinates of the multiple human body key points corresponding to the terminal posture of the target action is greater than the second displacement threshold, it indicates that the target human body has performed a target movement.

[0161] Exemplarily, in the above Example 1, a1 is taken as the first reference image of the first counting cycle, and a2~an are taken as the first alternative images. Starting from a2, a2 ​​is determined to be the current first alternative image; then based on the key point detection result of a1 and the key point detection result of a2, it is determined whether the displacement of the target human key points corresponding to the target motion in a2 is greater than or equal to the preset second displacement threshold; if not, a3 is taken as the new current first alternative image, and based on the key point detection result of a1 and the key point detection result of a3, it is determined whether the displacement of the target human key points corresponding to the target motion in a3 is greater than or equal to the preset second displacement threshold…, until a frame of the current first alternative image in which the displacement of the target human key points is greater than or equal to the preset second displacement threshold is found as the second reference image.

[0162] After the second reference image is successfully determined, the number of times the target movement is performed is updated. When the number of times the target movement is updated, for example, the current number of times the target movement is performed can be increased by a preset value, such as 1. In the first counting cycle, the current count of the target movement can be 0 or other non-zero values.

[0163] For the i-th counting cycle except the 1st counting cycle: an image whose timestamp is later than the second reference image determined in the i-1th counting cycle can be used as a second alternative image, and based on the key point detection results corresponding to the first reference image determined in the i-1th counting cycle of the target human body, and the key point detection results corresponding to the second alternative image, the first reference image corresponding to the i-th counting cycle is determined from the second alternative image; wherein i is an integer greater than 1.

[0164] For example, in the above example 1, the first reference image of the first cycle is determined to be a1. If the second reference image determined in the first counting cycle is a5, then after entering the second counting cycle, a6 to an are used as the second candidate images of the second counting cycle. Then, based on the key point detection result of a1 and the key point detection results corresponding to a6 to an, the first reference image corresponding to the first counting cycle is determined from a6 to an.

[0165] Exemplarily, when the second reference image corresponding to the i-th counting cycle is determined in this manner, for example, the following manner may be adopted:

[0166] A current second candidate image is determined from the second candidate images according to the sequence of timestamps.

[0167] Based on the key point detection result corresponding to the first reference image determined in the i-1th counting cycle of the target human body and the key point detection result corresponding to the current second alternative image, determine whether the displacement of the target human body key point corresponding to the target movement is less than a preset first displacement threshold.

[0168] In response to the displacement being less than the first displacement threshold, the current second candidate image is determined as the first reference image corresponding to the i-th counting cycle.

[0169] In response to the displacement being greater than or equal to the first displacement threshold, the process returns to the step of determining the current second candidate image from the second candidate images in the order of timestamps.

[0170] The first displacement threshold is the maximum value of the difference between the position information of multiple key points of the human body corresponding to the current posture of the target action and the position information of multiple key points of the human body corresponding to the starting posture of the target action. When the difference between the position information of multiple key points of the human body corresponding to the current posture of the target action and the position information of multiple key points of the human body corresponding to the starting posture of the target action is less than the first displacement threshold, it indicates that the target human body is resetting to the starting posture of the target action.

[0171] Exemplarily, in the above Example 1, a1 is used as the first reference image of the first counting cycle, and a6~an are used as the second alternative images corresponding to the second counting cycle. Starting from a6, a6 is determined as the current second alternative image; then based on the key point detection result of a1 and the key point detection result of a6, it is determined whether the displacement of the target human key points corresponding to the target motion in a6 is less than the first displacement threshold; if not, a7 is used as the new current second alternative image, and based on the key point detection result of a1 and the key point detection result of a7, it is determined whether the displacement of the target human key points corresponding to the target motion in a7 is less than the preset first displacement threshold..., until a frame of the current second alternative image in which the displacement of the target human key points is less than the preset first displacement threshold is found, which is used as the first reference image corresponding to the second counting cycle.

[0172] In this way, the number of times the target human body performs physical fitness tests can be counted more accurately, and the strength data obtained is also more accurate.

[0173] The embodiment of the present disclosure also provides a specific example of generating the force data of the target human body when performing the target movement, in which the following process is included:

[0174] First, the first reference image and the first candidate image are denoted as A1, a1, a2, a3, ..., a n , the key point coordinates corresponding to the target motion in the first reference image are marked as (x1, y1), and the key point coordinates corresponding to the target motion in each first candidate image are (x2, y2) (x3, y3) ... (x n ,y n ).

[0175] In the first counting cycle: A1 is the first reference image. At this time, a2, a3, ..., a n The second displacement threshold is D2. Then, according to the timestamp order, a2 is first used as the current first candidate image. According to (x1, y1), (x2, y2), the distance d2 between A1 and a2 is calculated. If the distance d2 is less than D2,

[0176] Then a3 is taken as the current first candidate image again, and the distance d3 between A1 and a3 is calculated according to (x1, y1) and (x3, y3). If the distance d3 is less than the preset distance threshold D2,

[0177] Then a4 is taken as the first candidate image, and the distance d4 between A1 and a4 is calculated based on (x1, y1) and (x4, y4). If the distance d4 is less than D2,

[0178] …

[0179] Then re-set a i As the current first candidate image, according to (x1, y1), (x i ,y i ), calculate A1, and a i The distance between i , if the distance d i is greater than or equal to D2, then a i The second candidate image corresponding to the terminal posture of completing a target action.

[0180] Increase the count by 1 and enter the second counting cycle:

[0181] A (i+1) ~a n Determined as the second candidate image.

[0182] Then, in order of timestamps, first (i+1) As the current second candidate image, since the first reference image of the first counting cycle is A1, when determining the first reference image of the second counting cycle, it is necessary to compare it with A1 because the action reset is required. The first displacement threshold is D1, then:

[0183] According to (x1, y1), (x (i+1) ,y (i+1) ), calculate A1, and a (i+1) The distance between (i+1) , if the distance d (i+1) Greater than D1,

[0184] Then re-set a (i+2) As the current first candidate image, according to (x1, y1), (x (i+2) ,y (i+2) ), calculate A1, and a (i+2) The distance between (i+2) , if the distance d (i+2) Greater than D1,

[0185] Then re-set a (i+3) As the current first candidate image, according to (x1, y1), (x (i+3) ,y (i+3) ), calculate A1, and a (i+3) The distance between (i+3) , if the distance d (i+3) Greater than D1,

[0186] …

[0187] Then re-set a jAs the current first candidate image, according to (x1, y1), (x j ,y j) , calculate A1, and a j The distance between j , if the distance d j If it is less than the preset distance threshold D1, then a j As the first reference image corresponding to the second counting cycle.

[0188] Then, the third counting cycle is entered, and actions similar to those of the second counting cycle are performed until all images in the motion video data are processed and the counting result is obtained.

[0189] Exemplarily, push-ups are taken as an example here, and the analysis method is also applicable to other sports as long as it is combined with the sports characteristics of the target sports.

[0190] The characteristic of push-up exercise is to keep the shoulder and neck moving up and down. Therefore, the displacement of the double shoulder points of the head and shoulder model can be counted.

[0191] In a specific implementation, when the push-up action begins, in the first reference image frame where the target movement starting posture is located, the starting coordinates of the shoulder points are (x, y), and the shoulder point coordinates (x1, y1) of the subsequent frame, i.e., the first candidate image, are continuously detected. n ,y n ), and calculate (x, y) and (x n ,y n ) (the shoulders will move up and down in the vertical direction during push-ups), and compare d with the counting threshold, i.e., the second displacement threshold D2. If it is greater than d, one count is completed, and then (xn, yn) is waited to be reset to the vicinity of the initial position. The second cycle counting is performed, and so on, and finally the movement number information of the target human body when performing the target movement is obtained.

[0192] Taking the specified time length of 1 minute as an example, the strength data for completing 20 sets of push-ups within 1 minute can be set to 100, and the strength data for completing 10 sets of push-ups within 1 minute is 50; based on the above method combined with the number of exercise information of the target human body when performing the target exercise, the strength data of the target human body when performing the target exercise is generated.

[0193] D: For the physical fitness test including a sensitivity test, the test result data includes the test result data corresponding to the sensitivity test of the target human body when performing the target movement: the target movement includes at least one of the following: curve dribbling, running around the pole, and cross quadrant jump.

[0194] In order to obtain the test result data corresponding to the sensitivity test, the key point detection results corresponding to the motion video data can be used to determine the second posture information of the target human body when performing the target motion.

[0195] The second posture information refers to the distance between the target limb and the marker in the target movement, and the body swing amplitude in the target movement.

[0196] Based on the similarity between the second posture information and the second target posture information corresponding to the target movement, posture validity information of the target human body when performing the target movement is obtained.

[0197] Based on the posture validity information of the target human body when performing the target movement, the movement duration of the target human body when performing the target movement, and the number of times the target movement is performed, the sensitivity information of the target human body when performing the target movement is determined.

[0198] The second target posture information refers to the distance between the target limb and the marker in the target movement, and the body swing amplitude in the target movement, which are preset as standard values.

[0199] In this way, the sensitivity information determined by using the posture validity information of the target human body when performing the target movement, as well as the movement duration and number of times, is more accurate.

[0200] In a specific implementation, when determining the sensitivity data of the target human body when performing the target motion, the following process may be performed:

[0201] Step 31: Based on the key point detection result, determine the second posture information of the target human body when performing the target movement; the second posture information includes at least one of the following: the distance between the target limb and the marker in the target movement, and the body swing amplitude in the target movement, for example: running around a pole, calculating the key points of the subject's limbs, and analyzing the body swing amplitude during the process of running around the pole, and the distance between the limbs and the upright pole when running around the pole (for example, the distance between the torso and the pole, the distance between the arms and the pole, and the distance between the hips and the pole).

[0202] Step 32: Based on the similarity between the second posture information and the second target posture information corresponding to the target movement, the posture validity information of the target human body when performing the target movement is obtained.

[0203] Step 33: Based on the posture validity information of the target human body when performing the target movement, the movement duration of the target human body when performing the target movement, and the number of times the target movement is performed, determine the sensitivity information of the target human body when performing the target movement.

[0204] In a specific implementation, for step 31, in a specific implementation, depending on the target movement type, a specific number of frames of motion images can be captured during the entire movement process, or the video can be directly recorded and multiple frames of motion images can be obtained; based on the average value of the distance and the body swing amplitude in the multiple frames of motion images, the distance between the limbs and the vertical pole and the body swing amplitude when the limbs are around the pole are obtained.

[0205] For example, in a pole running sport, a camera can be placed directly above the target body and ensure that its field of view covers the entire motion process, and the target body is filmed to obtain corresponding motion video data; another camera can be placed on the side of the target body and ensure that its field of view covers the entire motion process, and the target body is filmed to obtain corresponding motion video data.

[0206] The motion video data includes multiple frames of motion images. The multiple frames of motion images obtained by the camera directly above are used to determine the distance between the target limb and the marker in the target movement; the multiple frames of motion images obtained by the camera on the side are used to determine the body swing amplitude in the target movement.

[0207] In a specific implementation, the distance between the target limb and the marker in the target motion can be determined according to the method of determining the actual distance in the motion image in the speed test, and the body swing amplitude in the target motion can be determined according to the first posture information determination method in the endurance test. The similarities are not repeated here.

[0208] For step 32, in a specific implementation, the distance between the target limb in the second posture information and the marker in the target motion can be compared with the distance used as a standard, and the body swing amplitude in the target motion in the second posture information can be compared with the swing amplitude used as a standard to obtain the posture validity information of the target human body when performing the target motion.

[0209] Among them, the posture validity information is used to characterize the validity of the target human body when performing the target movement. For example, if the target movement is running around a pole, the smaller the distance between the target limb and the marker, and the smaller the body swing amplitude, the higher the validity degree; conversely, the larger the distance between the target limb and the marker, and the larger the body swing amplitude, the lower the validity degree.

[0210] With respect to step 33, in a specific implementation, the method for obtaining the number of movements in a strength test provided in an embodiment of the present disclosure may be used to obtain the number of times the target human body performs the target movement.

[0211] After obtaining the posture validity information, exercise duration and number of times the target human body performs the target movement, the sensitivity information of the target human body when performing the target movement is determined based on the different weights corresponding to the posture validity information, exercise duration and number of times in different target movement types.

[0212] E: The physical fitness test includes a flexibility test, and the test result data includes the test result data corresponding to the flexibility test of the target human body when performing the target exercise: the target exercise includes at least one of the following: high leg raises, yoga, and front sitting and lying postures.

[0213] In a specific implementation, when determining the flexibility data of the target human body when performing the target movement, the process described in the following steps 41 to 43 may be performed:

[0214] Step 41: Based on the key point detection result, determine the third posture information corresponding to the target limb in the target human body.

[0215] The third posture information refers to the position information of key points of the human body and the bending angle information of the target limbs; for example, in the sit-and-reach test shown, the bending angles of the torso and legs of the target human body.

[0216] Step 42: Compare the third posture information with the third target posture information corresponding to the target motion to determine the similarity between the third posture information and the third target posture information.

[0217] The third target posture information refers to the key point position information of the human body as a reference standard value, and the bending angle information of the target limb.

[0218] Step 43: Based on the similarity, generate flexibility data of the target human body when performing the target movement.

[0219] In this way, the flexibility information of the target human body when performing the target movement can be more accurately determined by utilizing the position information of the key points of the target human body when performing the target movement and the bending angle information of the target limbs.

[0220] In specific implementation, the difference between flexibility test and sensitivity test is that flexibility test pays more attention to the position information of key points of the human body and the bending angle information of the target limb, without adding factors such as time and distance. Therefore, the method for determining the position information of key points of the target limb and the bending angle information of the target limb in the flexibility test method is similar to that in the sensitivity test method, and will not be repeated here.

[0221] After determining the third posture information corresponding to the target limb in the target human body, the third posture information is compared with the third target posture information used as a standard in advance to determine the similarity between the third posture information and the third target posture information.

[0222] Among them, different target limbs of the target human body can be compared separately to determine the similarity between the third posture information and the third target posture information, or all target limbs of the target human body can be compared and the average value can be taken to determine the similarity between the third posture information and the third target posture information. Appropriate adjustments can be made based on different target movements and regulations.

[0223] For example, Figure 3 In the high-leg raising movement shown, the key points of the legs and the key points of the torso will present a specific angle. If the setting for the angle in the third target posture information is 135°, and the angle of the target human body performing the target movement obtained in the third posture information is 120°; then it is determined that the similarity between the third posture information and the third target posture information is approximately 89%.

[0224] After determining the similarity between the third posture information and the third target posture information, based on the similarity, the flexibility data of the target human body when performing the target movement is generated. For example, when the similarity is about 89%, the flexibility can be 89 between 0-100.

[0225] Those skilled in the art will appreciate that, in the above method of specific implementation, the order in which the steps are written does not imply a strict execution order and does not constitute any limitation on the implementation process. The specific execution order of the steps should be determined by their functions and possible internal logic.

[0226] Based on the same inventive concept, a physical fitness testing device corresponding to a physical fitness testing method is also provided in the embodiment of the present disclosure. Since the principle of solving the problem by the device in the embodiment of the present disclosure is similar to the above-mentioned physical fitness testing method in the embodiment of the present disclosure, the implementation of the device can refer to the implementation of the method, and the repeated parts will not be repeated.

[0227] Reference Figure 4 FIG. 4 is a schematic diagram of a physical fitness test device provided by an embodiment of the present disclosure, wherein the device comprises: an acquisition module 41, a detection module 42, and a generation module 43; wherein:

[0228] An acquisition module 41 is used to acquire motion video data of a target human body performing a target motion corresponding to the physical fitness test during a physical fitness test;

[0229] A detection module 42, configured to perform key point detection on the motion video data to obtain key point detection results corresponding to the motion video data;

[0230] The generation module 43 is used to generate test result data of the target human body when performing a physical fitness test based on the key point detection results; the test result data is used to reflect the physical fitness condition of the target human body in at least one measurement dimension.

[0231] In an optional embodiment, the physical fitness test includes at least one of the following: a speed test, an endurance test, a strength test, a sensitivity test, and a flexibility test.

[0232] In an optional implementation manner, the key point detection result includes: position information corresponding to a plurality of human key points in the target human body;

[0233] The generating module 43, when generating the test result data of the target human body when performing the physical fitness test based on the key point detection result, is used to:

[0234] Based on the key point detection result, determining the position information of the target human body key points corresponding to the target motion;

[0235] Based on the position information of the key points of the target human body, test result data of the target human body when performing a physical fitness test is generated.

[0236] In an optional implementation, the physical fitness test includes a speed test; the target exercise includes running;

[0237] The test result data corresponding to the speed test includes: speed information of the target human body when running within the target test range determined based on the running test distance; the speed information includes at least one of the following: the average moving speed, the maximum moving speed, the minimum moving speed, the first moving speed at the target movement moment, and the second moving speed at the target position when the running movement is performed within the target test range; wherein the target movement moment includes: any moment when the target human body performs running movement during the speed test; the target position includes: any position of the target human body within the target test range when the target human body performs running movement during the speed test.

[0238] In an optional implementation manner, the motion video data includes: multiple frames of motion images; the key point detection results, and the key point detection results corresponding to the multiple frames of the motion images respectively;

[0239] The generating module 43, when generating the test result data of the target human body when performing the physical fitness test based on the key point detection result, is used to:

[0240] Based on the key point detection results respectively corresponding to the multiple frames of the motion images, a target motion image corresponding to the speed information type is acquired from the multiple frames of the motion images;

[0241] The speed information is obtained based on a timestamp corresponding to the target motion image and a key point detection result corresponding to the target motion image.

[0242] In an optional embodiment, the method further includes at least one of the following:

[0243] In response to the speed information including: an average moving speed when the target motion is performed within the target test range, the target motion image includes: a first motion image of the target human body at a starting position corresponding to the target test range, and a second motion image of the target human body at an end position corresponding to the target test range;

[0244] In response to the speed information including: a first moving speed at a target moving moment, the target moving image includes: a third moving image in the moving video data whose timestamp is closest to the target moving moment, and a fourth moving image adjacent to the third moving image in terms of timestamp;

[0245] In response to the speed information including: a second moving speed of the target position, the target motion image includes: a fifth motion image in which the position of the target human body in the motion image in the motion video data is closest to the target position, and a sixth motion image adjacent to the fifth motion image in terms of timestamp; wherein the position of the target human body in the motion image is determined based on the key point detection result.

[0246] In an optional implementation manner, the generating module 43, in response to the speed information including: a maximum moving speed or a minimum moving speed when the target moves within the target test range, is configured to:

[0247] For each frame of the multiple frames of motion images, based on a key point detection result corresponding to each frame of the motion image and a key point detection result corresponding to a motion image that is adjacent to and / or adjacent to each frame of the motion image in terms of timestamp, determining a movement distance corresponding to each frame of the motion image;

[0248] Determine the instantaneous speed corresponding to each frame of the motion image based on the moving distance corresponding to each frame of the motion image, the timestamp corresponding to each frame of the motion image, and the timestamps corresponding to the adjacent motion images;

[0249] The maximum moving speed or the minimum moving speed is determined based on instantaneous speeds respectively corresponding to a plurality of frames of motion images.

[0250] In an optional implementation manner, the generating module 43, when determining the moving distance corresponding to each frame of the multiple frames of motion images based on the key point detection result corresponding to each frame of the motion image and the key point detection result corresponding to the motion image adjacent to the previous frame and / or adjacent to the next frame of the motion image in terms of timestamp, is used to:

[0251] Determine the displacement of the target human body in the image coordinate system corresponding to the image data based on the key point detection results corresponding to each frame of the motion image and the key point detection results corresponding to the motion image adjacent to and / or adjacent to each frame of the motion image in terms of timestamp;

[0252] Based on the proportional relationship between the predetermined image distance and the actual physical distance, and the displacement of the target human body in the image coordinate system corresponding to the image data, the moving distance corresponding to each frame of the motion image is determined.

[0253] In an optional implementation, the physical fitness test includes an endurance test; the test result data corresponding to the endurance test includes: endurance data of the target human body when performing the target exercise; the target exercise includes at least one of the following: long-distance running, rope skipping, and plank support;

[0254] The generating module 43, when generating the test result data of the target human body when performing the physical fitness test based on the key point detection result, is used to:

[0255] Determine, from the motion video data, target motion images corresponding to a plurality of motion stages, respectively; the plurality of motion stages respectively correspond to a plurality of time periods of the target motion; or respectively correspond to a plurality of sub-distances in the moving distance of the target motion;

[0256] For each motion stage, based on the key point detection result of the target motion image corresponding to the motion stage, determine the first posture information corresponding to the motion stage;

[0257] For each of the plurality of movement stages, matching the first posture information corresponding to the movement stage with the standard posture information to obtain the posture standard degree information corresponding to the movement stage of the target human body;

[0258] Determine the endurance data corresponding to the exercise stage based on the posture standard degree information and at least one of the exercise duration and the moving distance corresponding to the exercise stage;

[0259] Based on the endurance data corresponding to the multiple exercise stages, the endurance data of the target human body when performing the target exercise is obtained.

[0260] In an optional embodiment, the physical fitness test includes a strength test; the test result data corresponding to the strength test includes: strength data of the target human body when performing the target movement; the motion video data includes: multiple frames of motion images; the key point detection results include: key point detection results corresponding to the multiple frames of motion images; the target movement includes at least one of the following: pull-ups, push-ups, standing long jumps, and 10-minute supine leg raises;

[0261] The generating module 43, when generating the test result data of the target human body when performing the physical fitness test based on the key point detection result, is used to:

[0262] The following process is performed in each counting cycle of at least one counting cycle in the preset time length:

[0263] In response to the current counting cycle being the first counting cycle, determining the motion image with the earliest timestamp as the first reference image; in response to the current counting cycle being the i-th counting cycle, determining the motion image with a timestamp later than the second reference image determined in the i-1-th counting cycle as the second candidate image, and determining the first reference image corresponding to the i-th counting cycle from the second candidate images based on the key point detection result corresponding to the first reference image determined in the i-1-th counting cycle of the target human body and the key point detection result corresponding to the second candidate image; wherein i is an integer greater than 1;

[0264] Using other motion images whose timestamps are later than those of the first reference image as first candidate images, and determining a second reference image corresponding to a terminal posture when performing the target motion based on key point detection results corresponding to the first reference image and key point detection results corresponding to the candidate images;

[0265] In response to determining that the second reference image is successful, updating the number of times the target motion is performed;

[0266] In response to the end of the preset time period, based on the number of times the target body performs the target movement, the strength data of the target body when performing the target movement is determined; the strength value of the target body represented by the strength data is positively correlated with the number of times.

[0267] In an optional implementation, the physical fitness test includes a sensitivity test; the test result data corresponding to the sensitivity test includes: sensitivity information of the target human body when performing the target movement; the target movement includes at least one of the following: curve ball running, pole running, and cross quadrant jump;

[0268] The generating module 43, when generating the test result data of the target human body when performing the physical fitness test based on the key point detection result, is used to:

[0269] Based on the key point detection result, determine the second posture information of the target limb in the target human body during the target movement; the second posture information includes at least one of the following: the distance between the target limb and the marker in the target movement, and the body swing amplitude in the target movement;

[0270] Based on the similarity between the second posture information and the second target posture information corresponding to the target motion, obtaining posture validity information of the target human body when performing the target motion;

[0271] Based on the posture validity information of the target human body when performing the target movement, the movement duration of the target human body when performing the target movement, and the number of times the target movement is performed, the sensitivity information of the target human body when performing the target movement is determined.

[0272] In an optional embodiment, the physical fitness test includes a flexibility test; the test result data corresponding to the flexibility test includes: flexibility data of the target human body when performing the target exercise; the target exercise includes at least one of the following: high leg raises, yoga, and front-end sitting and lying;

[0273] The generating module 43, when generating the test result data of the target human body when performing the physical fitness test based on the key point detection result, is used to:

[0274] Based on the key point detection result, determine the third posture information corresponding to the target limb in the target human body; the third posture information includes at least one of the following: human body key point position information, and target limb bending angle information;

[0275] comparing the third posture information with the third target posture information corresponding to the target motion, and determining the similarity between the third posture information and the third target posture information;

[0276] Based on the similarity, flexibility data of the target human body when performing the target movement is generated.

[0277] For descriptions of the processing flow of each module in the device and the interaction flow between each module, reference may be made to the relevant descriptions in the above method embodiment, which will not be described in detail here.

[0278] The present disclosure also provides a computer device, such as Figure 5 FIG. 1 is a schematic diagram of a computer device structure provided in an embodiment of the present disclosure, including:

[0279] A processor 51 and a memory 52; the memory 52 stores machine-readable instructions executable by the processor 51, and the processor 51 is used to execute the machine-readable instructions stored in the memory 52. ​​When the machine-readable instructions are executed by the processor 51, the processor 51 performs the following steps:

[0280] Acquiring motion video data of a target human body performing a target motion corresponding to the physical fitness test during a physical fitness test;

[0281] Performing key point detection on the motion video data to obtain key point detection results corresponding to the motion video data;

[0282] Based on the key point detection results, test result data of the target person when undergoing a physical fitness test is generated, and the test result data is used to reflect the physical fitness condition of the target person in at least one measurement dimension.

[0283] The above-mentioned memory 52 includes an internal memory 521 and an external memory 522; the internal memory 521 here is also called an internal memory, which is used to temporarily store the calculation data in the processor 51, as well as the data exchanged with the external memory 522 such as a hard disk. The processor 51 exchanges data with the external memory 522 through the internal memory 521.

[0284] The specific execution process of the above instructions can refer to the steps of the physical fitness testing method described in the embodiment of the present disclosure, which will not be repeated here.

[0285] The present disclosure also provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the steps of the test method described in the above method embodiment are executed. The storage medium may be a volatile or non-volatile computer-readable storage medium.

[0286] The present disclosure also provides a computer program product that carries a program code. The program code includes instructions that can be used to execute the steps of the test method described in the above method embodiment. For details, please refer to the above method embodiment, which will not be repeated here.

[0287] The computer program product may be implemented in hardware, software or a combination thereof. In one optional embodiment, the computer program product is implemented as a computer storage medium. In another optional embodiment, the computer program product is implemented as a software product, such as a software development kit (SDK).

[0288] If the disclosed technical solution involves personal information, the product using the disclosed technical solution has clearly informed the personal information processing rules and obtained the individual's voluntary consent before processing the personal information. If the disclosed technical solution involves sensitive personal information, the product using the disclosed technical solution has obtained the individual's separate consent before processing the sensitive personal information, and at the same time meets the "explicit consent" requirement. For example, on personal information collection devices such as cameras, clear and prominent signs are set to inform that the personal information collection scope has been entered and personal information will be collected. If the individual voluntarily enters the collection scope, it is deemed that he or she agrees to the collection of his or her personal information; or on the device that processes personal information, when the personal information processing rules are notified by obvious signs / information, the individual's authorization is obtained through pop-up information or by asking the individual to upload his or her personal information; among them, the personal information processing rules may include information such as the personal information processor, the purpose of personal information processing, the processing method, and the type of personal information processed.

[0289] Those skilled in the art can clearly understand that, for the convenience and simplicity of description, the specific working process of the system and device described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here. In the several embodiments provided in the present disclosure, it should be understood that the disclosed system, device and method can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, and the indirect coupling or communication connection of the device or unit can be electrical, mechanical or other forms.

[0290] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0291] In addition, each functional unit in each embodiment of the present disclosure may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0292] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a non-volatile computer-readable storage medium that is executable by a processor. Based on this understanding, the technical solution of the present disclosure, 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, which is stored in a storage medium and includes several 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 disclosure. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0293] Finally, it should be noted that the above-described embodiments are only specific implementation methods of the present disclosure, which are used to illustrate the technical solutions of the present disclosure, rather than to limit them. The protection scope of the present disclosure is not limited thereto. Although the present disclosure is described in detail with reference to the above-described embodiments, ordinary technicians in the field should understand that any technician familiar with the technical field can still modify the technical solutions recorded in the above-described embodiments within the technical scope disclosed in the present disclosure, or can easily think of changes, or make equivalent replacements for some of the technical features therein; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present disclosure, and should be included in the protection scope of the present disclosure. Therefore, the protection scope of the present disclosure should be based on the protection scope of the claims.

Claims

1. A physical fitness test method, characterized in that: include: Acquiring motion video data of a target human body performing a target motion corresponding to the physical fitness test during a physical fitness test; Performing key point detection on the motion video data to obtain key point detection results corresponding to the motion video data; Based on the key point detection results, generating test result data of the target person when performing a physical fitness test, the test result data is used to reflect the physical fitness of the target person in at least one measurement dimension; The physical fitness test includes an endurance test; the test result data corresponding to the endurance test includes: endurance data of the target human body when performing the target exercise; the target exercise includes at least one of the following: long-distance running, rope skipping, and plank support; The generating, based on the key point detection result, test result data of the target human body when performing a physical fitness test comprises: Determine, from the motion video data, target motion images corresponding to a plurality of motion stages, respectively; the plurality of motion stages respectively correspond to a plurality of time periods of the target motion; or respectively correspond to a plurality of sub-distances in the moving distance of the target motion; For each motion stage, based on the key point detection result of the target motion image corresponding to the motion stage, determine the first posture information corresponding to the motion stage; For each of the plurality of movement stages, matching the first posture information corresponding to the movement stage with the standard posture information to obtain the posture standard degree information corresponding to the movement stage of the target human body; Determine the endurance data corresponding to the exercise stage based on the posture standard degree information and at least one of the exercise duration and the moving distance corresponding to the exercise stage; Based on the endurance data corresponding to the multiple exercise stages, the endurance data of the target human body when performing the target exercise is obtained.

2. The method according to claim 1, characterized in that: The physical fitness test also includes at least one of the following: a speed test, a strength test, a sensitivity test, and a flexibility test.

3. The method according to claim 1 or 2, characterized in that: The key point detection result includes: position information corresponding to a plurality of key points of the target human body; The generating, based on the key point detection result, test result data of the target human body when performing a physical fitness test comprises: Based on the key point detection result, determining the position information of the target human body key points corresponding to the target motion; Based on the position information of the key points of the target human body, test result data of the target human body when performing a physical fitness test is generated.

4. The method according to claim 1 or 2, characterized in that: The physical fitness test includes a speed test; the target exercise includes running; The test result data corresponding to the speed test includes: speed information of the target human body when running within the target test range determined based on the running test distance; the speed information includes at least one of the following: the average moving speed, the maximum moving speed, the minimum moving speed, the first moving speed at the target movement moment, and the second moving speed at the target position when the running movement is performed within the target test range; wherein the target movement moment includes: any moment when the target human body performs running movement during the speed test; the target position includes: any position of the target human body within the target test range when the target human body performs running movement during the speed test.

5. The method according to claim 4, characterized in that The motion video data includes: multiple frames of motion images; the key point detection results include: key point detection results corresponding to the multiple frames of motion images respectively; The generating, based on the key point detection result, test result data of the target human body when performing a physical fitness test comprises: Based on the key point detection results respectively corresponding to the multiple frames of the motion images, a target motion image corresponding to the speed information type is acquired from the multiple frames of the motion images; The speed information is obtained based on a timestamp corresponding to the target motion image and a key point detection result corresponding to the target motion image.

6. The method according to claim 5, characterized in that The method further comprises at least one of the following: In response to the speed information including: an average moving speed when the target motion is performed within the target test range, the target motion image includes: a first motion image of the target human body at a starting position corresponding to the target test range, and a second motion image of the target human body at an end position corresponding to the target test range; In response to the speed information including: a first moving speed at a target moving moment, the target moving image includes: a third moving image in the moving video data whose timestamp is closest to the target moving moment, and a fourth moving image adjacent to the third moving image in terms of timestamp; In response to the speed information including: a second moving speed of the target position, the target motion image includes: a fifth motion image in which the position of the target human body in the motion image in the motion video data is closest to the target position, and a sixth motion image adjacent to the fifth motion image in terms of timestamp; wherein the position of the target human body in the motion image is determined based on the key point detection result.

7. The method according to claim 4, characterized in that In response to the speed information including: a maximum moving speed or a minimum moving speed when the target moves within the target test range, including: For each frame of the multiple frames of motion images, based on a key point detection result corresponding to each frame of the motion image and a key point detection result corresponding to a motion image that is adjacent to and / or adjacent to each frame of the motion image in terms of timestamp, determining a movement distance corresponding to each frame of the motion image; Determine the instantaneous speed corresponding to each frame of the motion image based on the moving distance corresponding to each frame of the motion image, the timestamp corresponding to each frame of the motion image, and the timestamps corresponding to the adjacent motion images; The maximum moving speed or the minimum moving speed is determined based on instantaneous speeds respectively corresponding to a plurality of frames of motion images.

8. The method according to claim 7, characterized in that The determining, for each frame of the multiple frames of motion images, based on a key point detection result corresponding to each frame of the motion image and a key point detection result corresponding to a motion image that is adjacent to and / or adjacent to each frame of the motion image in terms of timestamp, a moving distance corresponding to each frame of the motion image comprises: Determine the displacement of the target human body in the image coordinate system corresponding to each frame of the motion image based on the key point detection result corresponding to each frame of the motion image and the key point detection result corresponding to the motion image adjacent to the previous frame and / or adjacent to the next frame of the motion image in terms of timestamp; Based on the proportional relationship between the predetermined image distance and the actual physical distance, and the displacement of the target human body in the image coordinate system corresponding to the motion image, the moving distance corresponding to each frame of the motion image is determined.

9. The method according to claim 1 or 2, characterized in that: The physical fitness test includes a strength test; The test result data corresponding to the strength test includes: the strength data of the target human body when performing the target movement; the movement video data includes: multiple frames of movement images; the key point detection results include: key point detection results corresponding to the multiple frames of movement images; the target movement includes at least one of the following: pull-ups, push-ups, standing long jumps, and 10-minute supine leg raises; The generating, based on the key point detection result, test result data of the target human body when performing a physical fitness test comprises: The following process is performed in each counting cycle of at least one counting cycle in the preset time length: In response to the current counting cycle being the first counting cycle, determining the motion image with the earliest timestamp as the first reference image; in response to the current counting cycle being the i-th counting cycle, determining the motion image with a timestamp later than the second reference image determined in the i-1-th counting cycle as the second candidate image, and determining the first reference image corresponding to the i-th counting cycle from the second candidate images based on the key point detection result corresponding to the first reference image determined in the i-1-th counting cycle of the target human body and the key point detection result corresponding to the second candidate image; wherein i is an integer greater than 1; Using other motion images whose timestamps are later than those of the first reference image as first candidate images, and determining a second reference image corresponding to a terminal posture when performing the target motion based on key point detection results corresponding to the first reference image and key point detection results corresponding to the candidate images; In response to determining that the second reference image is successful, updating the number of times the target motion is performed; In response to the end of the preset time period, based on the number of times the target body performs the target movement, the strength data of the target body when performing the target movement is determined; the strength value of the target body represented by the strength data is positively correlated with the number of times.

10. The method according to claim 1 or 2, characterized in that: The physical fitness test includes a sensitivity test; the test result data corresponding to the sensitivity test includes: sensitivity information of the target human body when performing the target movement; the target movement includes at least one of the following: curve ball running, pole running, and cross quadrant jump; The generating, based on the key point detection result, test result data of the target human body when performing a physical fitness test comprises: Based on the key point detection result, determine the second posture information of the target limb in the target human body during the target movement; the second posture information includes at least one of the following: the distance between the target limb and the marker in the target movement, and the body swing amplitude in the target movement; Based on the similarity between the second posture information and the second target posture information corresponding to the target motion, obtaining posture validity information of the target human body when performing the target motion; Based on the posture validity information of the target human body when performing the target movement, the movement duration of the target human body when performing the target movement, and the number of times the target movement is performed, the sensitivity information of the target human body when performing the target movement is determined.

11. The method according to claim 1 or 2, characterized in that: The physical fitness test includes a flexibility test; the test result data corresponding to the flexibility test includes: flexibility data of the target human body when performing the target exercise; the target exercise includes at least one of the following: high leg raises, yoga, and front-end sitting and lying; The generating, based on the key point detection result, test result data of the target human body when performing a physical fitness test comprises: Based on the key point detection result, determine the third posture information corresponding to the target limb in the target human body; the third posture information includes at least one of the following: human body key point position information, and target limb bending angle information; comparing the third posture information with the third target posture information corresponding to the target motion, and determining the similarity between the third posture information and the third target posture information; Based on the similarity, flexibility data of the target human body when performing the target movement is generated.

12. A physical fitness testing device, characterized in that: include: An acquisition module, used to acquire motion video data of a target human body performing a target motion corresponding to the physical fitness test during a physical fitness test; A detection module, used to perform key point detection on the motion video data to obtain key point detection results corresponding to the motion video data; A generating module, used for generating test result data of the target human body when performing a physical fitness test based on the key point detection result; The test result data is used to reflect the physical condition of the target human body in at least one measurement dimension; Wherein, the physical fitness test includes an endurance test; the test result data corresponding to the endurance test includes: endurance data of the target human body when performing the target exercise; the target exercise includes at least one of the following: long-distance running, rope skipping, and plank support; The generating module is specifically used for: Determine, from the motion video data, target motion images corresponding to a plurality of motion stages, respectively; the plurality of motion stages respectively correspond to a plurality of time periods of the target motion; or respectively correspond to a plurality of sub-distances in the moving distance of the target motion; For each motion stage, based on the key point detection result of the target motion image corresponding to the motion stage, determine the first posture information corresponding to the motion stage; For each of the plurality of movement stages, matching the first posture information corresponding to the movement stage with the standard posture information to obtain the posture standard degree information corresponding to the movement stage of the target human body; Determine the endurance data corresponding to the exercise stage based on the posture standard degree information and at least one of the exercise duration and the moving distance corresponding to the exercise stage; Based on the endurance data corresponding to the multiple exercise stages, the endurance data of the target human body when performing the target exercise is obtained.

13. A computer device, characterized in that: include: A processor and a memory, wherein the memory stores machine-readable instructions executable by the processor, and the processor is used to execute the machine-readable instructions stored in the memory. When the machine-readable instructions are executed by the processor, the processor executes the steps of the physical fitness testing method as described in any one of claims 1 to 11.

14. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program. When the computer program is executed by a computer device, the computer device executes the steps of the physical fitness testing method according to any one of claims 1 to 11.

Images