Sports guidance method and device based on computer vision and sports biomechanics difference, equipment and medium
By acquiring athletes' physical parameters and sports types, and combining this with biomechanical analysis to adjust scores, the problem of neglecting athlete differences in existing systems has been solved, resulting in more accurate and personalized scoring and guidance.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- SHANDONG PETROCHEMICAL INST
- Filing Date
- 2026-04-27
- Publication Date
- 2026-06-19
AI Technical Summary
Existing intelligent training systems ignore the differences between athletes when scoring movements and evaluating performance, resulting in inaccurate and untargeted scoring standards.
By acquiring athletes' physical parameters and exercise types based on computer vision, determining the standard range of physical parameters, analyzing real movements using a sports biomechanics knowledge base, and adjusting the initial movement score by combining physical parameter deviations and biomechanical information, a final movement score is obtained, and personalized exercise guidance is provided.
It improves the accuracy and relevance of scoring, provides personalized sports guidance reports, and helps athletes better understand their own differences and conduct targeted training.
Smart Images
Figure CN122244953A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of biomechanics, and in particular to methods, devices, equipment and media for motion guidance based on computer vision and differences in sports biomechanics. Background Technology
[0002] Currently, in sports training, the standardization of movement techniques and the efficiency of force application are the core metrics for measuring training quality. With the development of artificial intelligence and computer vision technology, real-time motion and posture recognition systems based on large AI models have been gradually applied in professional training. Among existing methods, some propose all-in-one machines that can recognize movements and provide posture standardization scores; others can perform high-precision three-dimensional posture analysis and violation judgment.
[0003] However, existing intelligent training systems still have limitations in terms of movement scoring and performance evaluation: their scoring criteria are usually based on a unified and fixed ideal movement model or standard athlete template. The system compares each athlete's real-time movement data with this universal template to obtain scores such as standardization and execution quality. This method ignores the differences between different athletes. Summary of the Invention
[0004] In view of this, the purpose of this invention is to provide a method, device, equipment, and medium for sports guidance based on computer vision and differences in sports biomechanics, which can fully take into account the differences between different athletes. The specific solution is as follows:
[0005] In the first aspect, this application discloses a motion guidance method based on computer vision and differences in sports biomechanics, including:
[0006] The target user's physical parameters are obtained using a visual camera, and the corresponding movement type of the target user is also obtained.
[0007] Based on the physical parameters, determine the standard physical parameter range for the target user under the exercise type, and based on the deviation between the physical parameters and the standard physical parameter range, determine the scoring adjustment factor;
[0008] The optimal sports biomechanical information of the target user under the sports type is determined by using a sports biomechanical knowledge base and based on the physical parameters. The actual movements of the target user are obtained based on the visual camera and inertial measurement unit sensor to determine the actual sports biomechanical information.
[0009] Based on the target optimal movement biomechanical information, the actual movement biomechanical information is analyzed to determine the initial movement score, and the initial movement score is adjusted according to the score adjustment factor to obtain the final movement score;
[0010] Based on the final movement score, the physical parameters, the standard physical parameter range, the target optimal exercise biomechanical information, and the actual exercise biomechanical information, an exercise guidance report is provided to the target user.
[0011] Optionally, the biomechanical information of the movement includes spatial completion in the spatial dimension, temporal rhythm in the temporal dimension, and movement technique efficiency in the dynamic dimension; the spatial completion includes joint angles and trajectory; the temporal rhythm includes the duration of each phase of the movement and the degree of matching with the standard rhythm model; the movement technique efficiency includes fluency, coordination, and energy efficiency.
[0012] Optionally, the step of analyzing the actual motion biomechanical information based on the target optimal motion biomechanical information to determine the initial movement score includes:
[0013] Based on the target optimal motion biomechanical information, the real motion biomechanical information is analyzed to determine the temporary motion scores corresponding to the spatial dimension, temporal dimension and dynamic dimension, respectively.
[0014] The initial action score is determined by weighted calculation based on the pre-assigned weights for different dimensions and the corresponding temporary action scores.
[0015] Optionally, determining the scoring adjustment factor based on the deviation between the physical parameters and the standard physical parameter range includes:
[0016] The target physical parameters corresponding to each target body shape are determined from the physical parameters, and the target parameter range corresponding to each target body shape is determined from the standard physical parameter range;
[0017] Calculate the target deviation between each target physical parameter and the corresponding target parameter range;
[0018] If the target deviation is not zero, then the corresponding initial adjustment factor is found from the preset deviation and factor mapping table corresponding to the target body shape to obtain several initial adjustment factors; the target body shape corresponds one-to-one with the preset deviation and factor mapping table.
[0019] Based on all the initial adjustment factors, the score adjustment factors are determined.
[0020] Optionally, determining the scoring adjustment factor based on all the initial adjustment factors includes:
[0021] Calculate the product of all the initial adjustment factors to obtain the score adjustment factor;
[0022] Accordingly, adjusting the initial action score according to the scoring adjustment factor to obtain the final action score includes:
[0023] The final action score is obtained by multiplying the scoring adjustment factor by the initial action score.
[0024] Optionally, calculating the target deviation between each target physical parameter and the corresponding target parameter range includes:
[0025] For each target physical parameter, if the target physical parameter is within the corresponding range of the target parameter, then zero is taken as the target deviation;
[0026] For each target physical parameter, if the target physical parameter is not within the corresponding range of the target parameters, the difference between each target physical parameter and the target parameter in the corresponding range of the target parameters is calculated, and the difference is used as the target deviation; the target parameter is the range boundary parameter that is closest to the target physical parameter within the range of the standard physical parameters.
[0027] Optionally, obtaining the movement type corresponding to the target user includes:
[0028] The exercise mode and exercise goal selected by the target user are obtained, and the exercise type is determined based on the exercise mode and exercise goal; the exercise type includes exercise posture type, dominant muscle group type and energy metabolism type.
[0029] Secondly, this application discloses a motion guidance device based on computer vision and differences in sports biomechanics, comprising:
[0030] The information acquisition module is used to acquire the physical parameters of the target user based on the visual camera, and to acquire the corresponding movement type of the target user;
[0031] The scoring adjustment factor determination module is used to determine the standard physical parameter range of the target user under the exercise type based on the physical parameters, and to determine the scoring adjustment factor based on the deviation between the physical parameters and the standard physical parameter range;
[0032] The exercise biomechanical information acquisition module is used to determine the target user's optimal exercise biomechanical information under the exercise type by using the exercise biomechanical knowledge base and based on the physical parameters, and to acquire the target user's real movements based on the visual camera and inertial measurement unit sensor to determine the real exercise biomechanical information.
[0033] The comprehensive scoring module is used to analyze the target user's actual motion biomechanical information based on the target's optimal motion biomechanical information to determine an initial motion score, and adjust the initial motion score according to the scoring adjustment factor to obtain a final motion score.
[0034] The exercise guidance module is used to provide an exercise guidance report to the target user based on the final movement score, the physical parameters, the standard physical parameter range, the target optimal exercise biomechanical information, and the actual exercise biomechanical information.
[0035] Thirdly, this application discloses an electronic device, including:
[0036] Memory, used to store computer programs;
[0037] A processor is configured to execute the computer program to implement the aforementioned disclosed motion guidance method based on computer vision and differences in sports biomechanics.
[0038] Fourthly, this application discloses a computer-readable storage medium for storing a computer program; wherein, when the computer program is executed by a processor, it implements the aforementioned motion guidance method based on differences in computer vision and sports biomechanics.
[0039] As can be seen, this application acquires the target user's physical parameters and corresponding exercise type based on a visual camera; determines the standard physical parameter range for the target user under the exercise type based on the physical parameters, and determines a scoring adjustment factor based on the deviation between the physical parameters and the standard physical parameter range; determines the target optimal exercise biomechanical information for the target user under the exercise type using an exercise biomechanical knowledge base and based on the physical parameters; acquires the target user's actual movements based on the visual camera and inertial measurement unit sensor to determine actual exercise biomechanical information; analyzes the actual exercise biomechanical information based on the target optimal exercise biomechanical information to determine an initial movement score, and adjusts the initial movement score according to the scoring adjustment factor to obtain a final movement score; and provides an exercise guidance report for the target user based on the final movement score, the physical parameters, the standard physical parameter range, the target optimal exercise biomechanical information, and the actual exercise biomechanical information. Therefore, this application takes into account the differences among different athletes, i.e., users, and proposes to adjust the initial movement score using different athletes' scoring adjustment factors, rather than simply scoring based on standard templates or ideal movements, thereby improving the accuracy and relevance of the scoring. This application determines the scoring adjustment factor by considering the deviation between the user's physical parameters and the standard physical parameter range, fully taking into account the physical characteristics of different users, thus improving the accuracy and relevance of the scoring. This application considers not only standard templates or ideal movements when scoring, but also sports biomechanical information to obtain the initial movement score, and then combines it with the scoring adjustment factor to obtain the final movement score. The use of sports biomechanical information here further improves the accuracy and relevance of the scoring and subsequent sports guidance reports. Attached Figure Description
[0040] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.
[0041] Figure 1 This application discloses a flowchart of a motion guidance method based on computer vision and differences in sports biomechanics.
[0042] Figure 2 This is a schematic diagram of a motion guidance device based on computer vision and differences in sports biomechanics disclosed in this application.
[0043] Figure 3 This is a structural diagram of an electronic device disclosed in this application. Detailed Implementation
[0044] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0045] Existing intelligent training systems still have limitations in terms of movement scoring and performance evaluation: their scoring standards are usually based on a unified and fixed ideal movement model or standard athlete template. The system compares each athlete's real-time movement data with this universal template to obtain scores such as standardization and execution quality. This method ignores the differences between different athletes.
[0046] Therefore, this application proposes a sports guidance scheme based on computer vision and sports biomechanics differences, which can fully take into account the differences between different athletes.
[0047] This application discloses a motion guidance method based on computer vision and differences in sports biomechanics. (See also...) Figure 1 As shown, the method includes:
[0048] Step S11: Obtain the physical parameters of the target user based on the visual camera, and obtain the corresponding movement type of the target user.
[0049] In this embodiment, the physical parameters include, but are not limited to, height (H), arm span (A), leg length (L), shoulder width (S), and foot length (F). This application acquires physical information through visual calibration (guiding the user to perform standard standing postures and strides) or in combination with simple measurement tools (visual cameras, binocular cameras, depth sensors). The collected data can be uploaded to the cloud to generate a personalized biometric profile for the user. It should be noted that if the target user is using the exercise guidance system for the first time, physical parameters need to be collected temporarily. If it is not the first time using the system, any previously collected physical parameters can be selected, or temporary parameters can be collected.
[0050] In one specific embodiment, the user stands in front of a device equipped with a visual calibration module, completes a standardized posture according to on-screen instructions, and the system measures key body parameters using a binocular camera or depth sensor. The standardized posture includes, but is not limited to, a static standing posture and a stepping posture. The static standing posture involves standing barefoot naturally. The system automatically calculates the user's height (H=178cm), arm span (A=168cm (significantly shorter than height, indicating a short arm span), shoulder width (S=38cm), leg length (L=85cm), and foot length (F=26cm) using a binocular camera or depth sensor. Other parameters are not specifically exemplified here. The stepping posture involves performing a slight squat. The system records the initial range of motion of the hip, knee, and ankle joints and inquires about the user's dominant hand (e.g., right-handed).
[0051] In this embodiment, obtaining the exercise type corresponding to the target user includes: obtaining the exercise mode and exercise goal selected by the target user, and determining the exercise type based on the exercise mode and the exercise goal; the exercise type includes exercise posture type, dominant muscle group type and energy metabolism type.
[0052] It should be noted that the exercise mode includes exercises such as pull-ups and standing long jump; the exercise goals include exercise indicators such as quantity (e.g., 10 per minute), quality (standard of movement), and distance; the exercise posture type includes supine and standing postures; the dominant muscle group type includes upper limbs and lower limbs; and the energy metabolism type includes strength and endurance.
[0053] In one specific embodiment, exercise classification matching is performed based on the exercise mode as pull-ups and the goal as number of repetitions / endurance: Posture type: standing (vertical suspension); Dominant muscle group type: upper limb (back and arms are the main muscle groups); Energy metabolism type: endurance (aiming to complete multiple repetitions). Exercise type is determined as: [standing, upper limb, endurance].
[0054] Step S12: Determine the standard physical parameter range of the target user under the exercise type based on the physical parameters, and determine the scoring adjustment factor based on the deviation between the physical parameters and the standard physical parameter range.
[0055] It should be noted that the standard range of the same physical parameter differs across different types of exercise. For example, when the exercise includes standing movements, the range of standard parameters for leg length will be narrowed (emphasizing the influence of the legs on the movement), while the range of standard parameters for arm length will be expanded (weakening the influence of the arms on the movement).
[0056] In this embodiment, determining the standard physical parameter range for the target user under the exercise type based on the physical parameters can be achieved by calculating the standard range of other physical parameters based on height, or by calculating the required standard range of physical parameters based on the exercise type, while omitting the calculation of unnecessary parameters. To ensure timely use later, the standard range of all physical parameters can also be calculated. It should be noted that other physical parameters can be determined by multiplying height by a preset value, or other methods can be used; no specific limitation is made here.
[0057] In one specific embodiment, the system retrieves the standard physical parameter ranges optimized for this type of exercise from the physical parameter knowledge base based on the type [standing, upper limb, endurance]. (It should be noted that arm span and shoulder width are retained here, while leg length and other lower limb parameters are not considered). For exercises that focus on upper limbs and endurance, the system strengthens the influence weight of the arm span parameter, setting its standard range relatively loosely to accommodate the physiological structure of most people, but highlighting its importance. Specifically, the standard arm span range is [height × 0.98, height × 1.05], that is, for an ideal model with a height of 178cm, the standard arm span is approximately [174.4 cm, 186.9 cm], and the standard shoulder width range is [36 cm, 42 cm].
[0058] In this embodiment, determining the scoring adjustment factor based on the deviation between the physical parameters and the standard physical parameter range includes: determining the target physical parameters corresponding to each target body shape from the physical parameters, and determining the target parameter range corresponding to each target body shape from the standard physical parameter range; calculating the target deviation between each target physical parameter and the corresponding target parameter range; if the target deviation is not zero, finding the corresponding initial adjustment factor from the preset deviation and factor mapping table corresponding to the target body shape to obtain several initial adjustment factors; the target body shape corresponds one-to-one with the preset deviation and factor mapping table; and determining the scoring adjustment factor based on all the initial adjustment factors.
[0059] It should be noted that the greater the target deviation, the greater the difficulty for the user to complete the action, and the higher the score for this item will be (based on the adjustment factor to improve the score), allowing users to avoid poor athletic performance due to their own physical parameters, and thus train more effectively.
[0060] In this embodiment, determining the score adjustment factor based on all the initial adjustment factors includes: calculating the product of all the initial adjustment factors to obtain the score adjustment factor.
[0061] It should be noted that not all physical parameters are outside the standard target parameter range. Only those outside the corresponding range need to be adjusted, that is, there will be corresponding adjustment factors. If no adjustment is needed, there will be no adjustment factor (that is, there is no adjustment factor when the target deviation is zero). This is because there may be several adjustment factors. When the number is greater than one (for example, there are two physical parameters, arm span and shoulder width, and the deviations of both are not zero, so there are two initial adjustment factors), it is necessary to count all initial adjustment factors to obtain the scoring adjustment factor. The specific adjustment method can be to calculate the product of all initial adjustment factors, sum them, etc.
[0062] In this embodiment, calculating the target deviation between each target physical parameter and the corresponding target parameter range includes: for each target physical parameter, if the target physical parameter is within the corresponding target parameter range, then zero is used as the target deviation; for each target physical parameter, if the target physical parameter is not within the corresponding target parameter range, then the difference between each target physical parameter and the target parameter in the corresponding target parameter range is calculated, and the difference is used as the target deviation; the target parameter is the range boundary parameter within the standard physical parameter range that is closest to the target physical parameter.
[0063] In one specific embodiment, the standard arm span is approximately [174.4cm, 186.9cm]. The user's arm span A = 168cm, which is lower than the lower limit of the standard range of 174.4cm. The user's arm span is closest to the lower limit of the standard arm span range, and the calculated target deviation is |168-174.4| = 6.4cm. The target deviation is not zero, and there is a corresponding initial adjustment factor. The standard shoulder width range is [36cm, 42cm]. The shoulder width S = 38cm, which falls within the standard range [36cm, 42cm]. The target deviation is 0, and there is no initial adjustment factor in this case. According to the preset deviation and factor mapping table corresponding to the arm span, the initial adjustment factor corresponding to 6.4cm is determined to be 1.08. Since there is no corresponding initial adjustment factor for the shoulder width, the final adjustment factor in this specific embodiment is the initial adjustment factor corresponding to 6.4cm, which is 1.08. If the shoulder width is not within the standard shoulder width range, then there is an initial adjustment factor. In this case, the final adjustment factor is the product of the initial adjustment factor corresponding to the shoulder width and the initial adjustment factor corresponding to the arm span.
[0064] Step S13: Determine the target user's optimal sports biomechanical information under the sports type by using the sports biomechanical knowledge base and based on the physical parameters, and obtain the target user's real movements based on the visual camera and inertial measurement unit sensor to determine the real sports biomechanical information.
[0065] It should be noted that the optimal biomechanical information for the target movement can be range information, such as the range of joint angles.
[0066] In this embodiment, the biomechanical information of exercise includes spatial completion in the spatial dimension, temporal rhythm in the temporal dimension, and movement technique efficiency in the dynamic dimension; the spatial completion includes joint angles and trajectory; the temporal rhythm includes the duration of each stage of the movement and the degree of matching with the standard rhythm model; the movement technique efficiency includes fluency, coordination, and energy efficiency.
[0067] It should be noted that visual cameras are used to capture full-body motion videos, such as multi-view cameras simultaneously capturing full-body motion videos. Inertial measurement unit sensors can be worn at the locations where measurements are needed to record data such as swing acceleration and angular velocity. All data is synchronized and preprocessed in real time at the edge. Edge computing units (devices) fuse multi-source data, reconstruct the 3D sequence of motion, and perform frame-by-frame and dimension-by-dimensional comparative analysis.
[0068] In one specific embodiment, for the case where the exercise mode is pull-ups and the goal is quantity / endurance (i.e., [standing, upper limb, endurance] type), firstly, the optimal sports biomechanical information for the target is selected from the sports biomechanical knowledge base. Subsequently, the actual spatial completion of the pull-up is determined based on the actual movement, that is, the limb posture, movement trajectory, joint angles, etc. are determined; the temporal rhythm of the pull-up is determined based on the actual movement, that is, the time allocation, movement frequency and periodicity of each stage of the movement (ascent, pause, descent), etc. are determined; the movement technique efficiency of the pull-up is determined based on the actual movement, that is, the ability to complete an effective movement per unit time, or the timeliness, continuity and efficiency of the movement, as well as the coordination and symmetry of the movement, such as the coordination and symmetry of the arm trajectory: whether the arm swing amplitude is different, whether there is asymmetry in force exertion; whether the body rotates, whether the core is insufficiently engaged, etc.
[0069] In another specific embodiment, for the case where the movement mode is standing long jump and the movement target is mass (i.e., the movement type is [standing, lower limb, strength]), the optimal biomechanical information of the target includes: the optimal take-off angle range: based on his leg length and center of gravity height, the most suitable take-off angle is calculated as follows: - (rather than general) - Ideal leg tuck angle during takeoff: Based on the leg length and biomechanical efficiency model, the optimal knee folding angle during takeoff is... - Ideal landing posture: Based on height and foot length, calculate the recommended angle range between the lower leg and the ground upon landing. Real-world biomechanical information includes: Spatial completion: Takeoff angle. (exist - Within the acceptable range (excellent); knee joint angle at the highest point of the jump (exist - Within acceptable limits; the lower leg angle upon landing was slightly excessive. Timing and rhythm: Analysis of the time proportions of the takeoff, flight, and landing cushioning phases revealed that Zhang San's flight time was slightly short, while his landing cushioning time was too long, indicating insufficient force concentration and a need for improvement in landing technique. Dynamics: Analysis of the symmetry and coordination of the arm swing trajectory using IMU (inertial measurement unit) data revealed that his right arm swing amplitude was greater than his left arm, indicating asymmetrical force exertion; further analysis using motion video can be helpful. A slight leftward rotation of the body in the air indicates insufficient core engagement.
[0070] Step S14: Based on the target optimal motion biomechanical information, analyze the real motion biomechanical information to determine the initial motion score, and adjust the initial motion score according to the score adjustment factor to obtain the final motion score.
[0071] It should be noted that the specific scoring process can be completed by edge computing devices.
[0072] It should be noted that the optimal biomechanical information for the target movement can be range information, while the actual biomechanical information is a numerical value. If the value is within the range, the score is high; if the value is outside the range, the score decreases as the distance from the range increases. Specifically, this can be achieved by setting a mapping relationship between the preset score and the target for each type of biomechanical information in each dimension. The target relationship is the relationship between the numerical value and the range, including both within and outside the range, and the distance from the range boundary outside the range (the farther from the range boundary, the lower the score). Scores within the range can be consistent, and scores within the range can also be adjusted based on the relationship with the range boundaries (the farther from both range boundaries, the higher the score). If multiple scores exist within the same dimension, a temporary movement score can be calculated using a weighted algorithm, and the weights can be preset according to the actual situation.
[0073] In one specific embodiment, 100 points can be the full score. If the value is not within the corresponding range, points will be deducted. The deduction standard can be obtained based on the pre-defined mapping relationship between the deduction score and the distance for each type of biomechanical information under each dimension. If the value is within the range, the score remains unchanged. If the value is outside the range, the further away from the boundary of the range, the more points will be deducted. The specific deduction situation can be set according to the actual situation, which will not be specifically described here.
[0074] In this embodiment, there is multi-dimensional motion biomechanical information. Therefore, when determining the initial action score, it is necessary to determine the temporary action scores for each dimension, and then weight them to obtain the initial action score. Specifically, the step of analyzing the real motion biomechanical information based on the target optimal motion biomechanical information to determine the initial action score includes: analyzing the real motion biomechanical information based on the target optimal motion biomechanical information to determine the temporary action scores corresponding to the spatial dimension, temporal dimension, and dynamic dimension, respectively; and performing a weighted calculation based on the weights pre-assigned to different dimensions and the corresponding temporary action scores to determine the initial action score.
[0075] In one specific embodiment, the weights for the spatial dimension are set to 0.4, the weights for the temporal dimension are set to 0.3, and the weights for the dynamic dimension are set to 0.4.
[0076] In one specific embodiment, for the case where the exercise mode is pull-ups and the goal is quantity / endurance (i.e., [standing, upper limb, endurance] type), firstly, the optimal exercise biomechanical information for the target is selected from the exercise biomechanical knowledge base. Subsequently, the actual spatial completion of the pull-up is determined based on the actual movement, that is, the limb posture, movement trajectory, joint angles, etc., are determined, and a temporary movement score of spatial completion is obtained by comparing the actual spatial completion with the optimal spatial completion of the target. Based on the actual movement, the temporal rhythm of the pull-up is determined, that is, the time allocation of each stage of the movement (ascent, pause, descent), movement frequency and periodicity, etc., are determined, and compared with the actual movement. A provisional movement score for temporal rhythm is obtained by comparing the actual temporal rhythm with the target optimal temporal rhythm. Based on the actual movement, the movement technique efficiency of the pull-up is determined, which is the ability to complete an effective movement per unit time, or the timeliness, continuity, and efficiency of the movement, as well as the coordination and symmetry of the movement, such as the coordination and symmetry of the arm trajectory: whether the arm swing amplitude is different, whether there is asymmetry in force exertion; whether the body rotates, whether the core is insufficiently engaged, etc. A provisional movement score for movement technique efficiency is obtained by comparing the actual movement technique efficiency with the target optimal movement technique efficiency. The three provisional movement scores are weighted and calculated to determine the initial movement score. Assuming a spatial dimension score of 85 (out of 100), a temporal dimension score of 70 (out of 100), and a dynamic dimension score of 60 (out of 100), the initial movement score = 85 × 0.4 + 70 × 0.3 + 60 × 0.3 = 73.
[0077] In another specific embodiment, for the case where the movement mode is standing long jump and the movement goal is mass (that is, the movement type is [standing, lower limb, strength]), the spatial dimension is 85 points (out of 100), the temporal dimension is 70 points (out of 100), and the dynamic dimension is 65 points (out of 100). The initial movement score is 85×0.4+70×0.3+65×0.3=76.
[0078] It should be noted that, in addition to calculating the temporary action score, the initial action score, and the final action score, it is also possible to calculate the product of the temporary action score and the score adjustment factor for each dimension to obtain the adjusted score. The adjusted score can be displayed in the form of a multi-dimensional radar chart in the subsequent report.
[0079] In one specific embodiment, the temporary action score is: 85 points (out of 100) for the spatial dimension, 70 points (out of 100) for the temporal dimension, and 60 points (out of 100) for the dynamic dimension; the adjusted score is: 85 points × 1.08 = 91.8 for the spatial dimension, 70 points × 1.08 = 75.6 for the temporal dimension, and 60 points × 1.08 = 64.8 for the dynamic dimension.
[0080] In this embodiment, adjusting the initial action score according to the scoring adjustment factor to obtain the final action score includes: calculating the product of the scoring adjustment factor and the initial action score to obtain the final action score.
[0081] In one specific embodiment, for the case where the exercise mode is pull-ups and the goal is quantity / endurance (i.e., the type of [standing, upper limb, endurance]), the initial movement score is 73, the score adjustment factor is 1.08, and the final movement score is 73 × 1.08 = 78.84.
[0082] Step S15: Based on the final action score, the physical parameters, the standard physical parameter range, the target optimal exercise biomechanical information, and the actual exercise biomechanical information, provide an exercise guidance report to the target user.
[0083] It should be noted that the exercise guidance report includes a score display, physical problems, exercise problems corresponding to biomechanical information, and training programs.
[0084] It's important to note that the rating display includes not only the final action score but also a multi-dimensional radar chart, clearly showing scores in three dimensions: spatial completion, tempo, and action technique effectiveness (these scores can be adjusted). This allows users to immediately identify their biggest weaknesses. The rating results are not presented as a single total score but rather as a multi-dimensional radar chart or bar graph showing the user's scores across each dimension. Users can easily see their strengths (such as good tempo) and weaknesses (such as poor spatial completion).
[0085] It's important to note that when a score in a particular dimension is low, the system can drill down further. For example, if the spatial completion score is low, the system will pinpoint the specific joint and mark the specific frame on the action timeline. By overlaying AR (Augmented Reality) annotations (such as highlighting the problematic joint and drawing the ideal trajectory line) onto the action playback video, the system can visually indicate the problem. The guidance report should also include a textual diagnostic summary, such as: Your takeoff angle is excellent, but your air posture control and force symmetry need improvement, leading to power loss and affecting jump distance. The guidance report can also include links to relevant training courses.
[0086] It should be noted that the exercise guidance report may include an individual exercise curve, which not only shows the changes in the total score, but also the evolution of abilities in each dimension, enhancing the motivation to continue using the system and trust in it.
[0087] In one specific embodiment, for the exercise mode of pull-ups and the goal of quantity / endurance (i.e., [standing, upper limb, endurance]), the exercise guidance report content is as follows: Scoring display: "Pull-up score: 78.84 points (adjusted for fairness based on your physical condition)", with adjusted scores for each dimension as follows: Spatial dimension 91.8 points, Temporal dimension 75.6 points, Kinetic dimension 64.8 points. Diagnostic interpretation: First, it states: "Based on assessment, your arm span condition increases the physiological difficulty of completing a standard pull-up (individualized difficulty coefficient has been included in the score)." Then, it explains the technical issues: "During the movement, it was observed that you experienced swing compensation and inconsistent force exertion in subsequent repetitions. The following modifications are recommended: (specific suggestions are explained)." Training prescription: The generated training suggestions will take into account your physical characteristics, such as: strengthening the static hold of vertical suspension, enhancing grip and latissimus dorsi activation to compensate for the relatively high pull-up challenge; performing eccentric control training to improve movement control, etc.
[0088] In one specific embodiment, for the exercise mode of pull-ups and the goal of quantity / endurance (i.e., [standing, upper limb, endurance] type), the exercise guidance report includes the following: final movement score; multi-dimensional radar chart: clearly showing the scores (adjusted scores) of three dimensions: spatial completion, temporal rhythm, and movement technique effectiveness, making it immediately clear that movement technique effectiveness is the biggest weakness. Problem frame location and AR playback: clicking on the movement technique effectiveness dimension, the system automatically jumps to the keyframes of "leftward body rotation in the air" and "asymmetrical arm swing" in the movement video, and uses AR arrows and curves to overlay on the screen, intuitively pointing out the problem. Text diagnostic summary: "Your take-off angle is excellent, but your in-air posture control and force symmetry need to be strengthened, resulting in power loss and affecting the jump distance." Targeted corrective training plan: Training 1 (improving arm swing symmetry): stationary arm swing practice, facing a mirror, focusing on the synchronous swing of both arms, 3 sets × 20 repetitions. Training 2 (strengthening core control): supine leg raises or plank to improve body stability in the air, 3 sets × 30 seconds. It is recommended to perform the above-mentioned targeted training twice a week; related micro-course learning: two short video links will be automatically pushed: "How to achieve symmetrical arm swing and power" and "Detailed explanation of the standing long jump mid-air abdominal contraction movement".
[0089] It should be noted that the guidance report in this application is not merely an abstract result such as "jumped 2 meters and scored 76 points" or "did 10 pull-ups and scored 78.84 points", but is transformed into a clear, credible and actionable guidance plan that includes scoring, specific sports problems and training programs, thus achieving a leap from simply scoring to replacing some of the coach's work.
[0090] As can be seen, this application acquires the target user's physical parameters and corresponding exercise type based on a visual camera; determines the standard physical parameter range for the target user under the exercise type based on the physical parameters, and determines a scoring adjustment factor based on the deviation between the physical parameters and the standard physical parameter range; determines the target optimal exercise biomechanical information for the target user under the exercise type using an exercise biomechanical knowledge base and based on the physical parameters; acquires the target user's actual movements based on the visual camera and inertial measurement unit sensor to determine actual exercise biomechanical information; analyzes the actual exercise biomechanical information based on the target optimal exercise biomechanical information to determine an initial movement score, and adjusts the initial movement score according to the scoring adjustment factor to obtain a final movement score; and provides an exercise guidance report for the target user based on the final movement score, the physical parameters, the standard physical parameter range, the target optimal exercise biomechanical information, and the actual exercise biomechanical information. Therefore, this application takes into account the differences among different athletes, i.e., users, and proposes to adjust the initial movement score using different athletes' scoring adjustment factors, rather than simply scoring based on standard templates or ideal movements, thereby improving the accuracy and relevance of the scoring. This application determines the scoring adjustment factor by considering the deviation between the user's physical parameters and the standard physical parameter range, fully taking into account the physical characteristics of different users, thus improving the accuracy and relevance of the scoring. This application considers not only standard templates or ideal movements when scoring, but also sports biomechanical information to obtain the initial movement score, and then combines it with the scoring adjustment factor to obtain the final movement score. The use of sports biomechanical information here further improves the accuracy and relevance of the scoring and subsequent sports guidance reports.
[0091] Accordingly, embodiments of this application also disclose a motion guidance device based on computer vision and differences in sports biomechanics, see [link to relevant documentation]. Figure 2 As shown, the device includes:
[0092] The information acquisition module 11 is used to acquire the physical parameters of the target user based on the visual camera, and to acquire the corresponding movement type of the target user;
[0093] The scoring adjustment factor determination module 12 is used to determine the standard physical parameter range of the target user under the exercise type based on the physical parameters, and to determine the scoring adjustment factor based on the deviation between the physical parameters and the standard physical parameter range;
[0094] The sports biomechanical information acquisition module 13 is used to determine the target optimal sports biomechanical information of the target user under the sports type by means of the sports biomechanical knowledge base and based on the body parameters, and to acquire the actual movements of the target user based on the visual camera and the inertial measurement unit sensor to determine the actual sports biomechanical information.
[0095] The comprehensive scoring module 14 is used to analyze the actual motion biomechanical information of the target user based on the target's optimal motion biomechanical information to determine the initial motion score, and adjust the initial motion score according to the scoring adjustment factor to obtain the final motion score.
[0096] The exercise guidance module 15 is used to provide an exercise guidance report to the target user based on the final movement score, the physical parameters, the standard physical parameter range, the target optimal exercise biomechanical information, and the actual exercise biomechanical information.
[0097] The more specific working process of each of the above modules can be found in the corresponding content disclosed in the foregoing embodiments, and will not be repeated here.
[0098] Furthermore, embodiments of this application also provide an electronic device. Figure 3 This is a structural diagram of an electronic device 20 according to an exemplary embodiment. The content of the diagram should not be construed as limiting the scope of this application.
[0099] Figure 3 This is a schematic diagram of the structure of an electronic device 20 provided in an embodiment of this application. Specifically, the electronic device 20 may include: at least one processor 21, at least one memory 22, a display screen 23, an input / output interface 24, a communication interface 25, a power supply 26, and a communication bus 27. The memory 22 stores a computer program, which is loaded and executed by the processor 21 to implement the relevant steps in the motion guidance method based on computer vision and biomechanical differences disclosed in any of the foregoing embodiments. Furthermore, the electronic device 20 in this embodiment may specifically be an electronic computer.
[0100] In this embodiment, the power supply 26 is used to provide operating voltage for each hardware device on the electronic device 20; the communication interface 25 can create a data transmission channel between the electronic device 20 and external devices, and the communication protocol it follows can be any communication protocol applicable to the technical solution of this application, and is not specifically limited here; the input / output interface 24 is used to acquire external input data or output data to the outside world, and its specific interface type can be selected according to specific application needs, and is not specifically limited here.
[0101] Furthermore, the memory 22, as a carrier for resource storage, can be a read-only memory, random access memory, disk, or optical disk, etc. The resources stored thereon may include computer programs 221, and the storage method may be temporary storage or permanent storage. The computer programs 221 may include, in addition to computer programs capable of performing the motion guidance method based on computer vision and biomechanical differences, executed by the electronic device 20 as disclosed in any of the foregoing embodiments, computer programs capable of performing other specific tasks.
[0102] Furthermore, embodiments of this application also disclose a computer-readable storage medium for storing a computer program; wherein, when the computer program is executed by a processor, it implements the aforementioned motion guidance method based on differences in computer vision and sports biomechanics.
[0103] The specific steps of this method can be found in the corresponding content disclosed in the foregoing embodiments, and will not be repeated here.
[0104] The various embodiments in this application are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. For the same or similar parts between the various embodiments, refer to each other. As for the apparatus disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple, and relevant parts can be referred to in the method section.
[0105] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.
[0106] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly by hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.
[0107] Finally, it should be noted that in this document, relational terms such as "first" and "first" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.
[0108] The foregoing has provided a detailed description of a motion guidance method, apparatus, device, and storage medium based on differences in computer vision and sports biomechanics provided in this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.
Claims
1. A method of sports coaching based on computer vision and sports biomechanics differences, characterized in that, include: The target user's physical parameters are obtained using a visual camera, and the corresponding movement type of the target user is also obtained. Based on the physical parameters, determine the standard physical parameter range for the target user under the exercise type, and based on the deviation between the physical parameters and the standard physical parameter range, determine the scoring adjustment factor; The optimal sports biomechanical information of the target user under the sports type is determined by using a sports biomechanical knowledge base and based on the physical parameters. The actual movements of the target user are obtained based on the visual camera and inertial measurement unit sensor to determine the actual sports biomechanical information. Based on the target optimal movement biomechanical information, the actual movement biomechanical information is analyzed to determine the initial movement score, and the initial movement score is adjusted according to the score adjustment factor to obtain the final movement score; Based on the final movement score, the physical parameters, the standard physical parameter range, the target optimal exercise biomechanical information, and the actual exercise biomechanical information, an exercise guidance report is provided to the target user.
2. The method of claim 1, wherein the method is based on computer vision and motion biomechanics differences. Sports biomechanical information includes spatial completion in the spatial dimension, temporal rhythm in the temporal dimension, and movement technique efficiency in the dynamic dimension; the spatial completion includes joint angles and trajectory; the temporal rhythm includes the duration of each phase of movement and the degree of matching with the standard rhythm model; the movement technique efficiency includes fluency, coordination, and energy efficiency.
3. The method of coaching sports based on computer vision and sports biomechanics differences as claimed in claim 2, wherein, The process of analyzing the actual motion biomechanical information based on the target optimal motion biomechanical information to determine the initial movement score includes: Based on the target optimal motion biomechanical information, the real motion biomechanical information is analyzed to determine the temporary motion scores corresponding to the spatial dimension, temporal dimension and dynamic dimension, respectively. The initial action score is determined by weighted calculation based on the pre-assigned weights for different dimensions and the corresponding temporary action scores.
4. The method of coaching sports based on computer vision and sports biomechanics differences as claimed in claim 1, wherein, The determination of the scoring adjustment factor based on the deviation between the physical parameters and the standard physical parameter range includes: The target physical parameters corresponding to each target body shape are determined from the physical parameters, and the target parameter range corresponding to each target body shape is determined from the standard physical parameter range; Calculate the target deviation between each target physical parameter and the corresponding target parameter range; If the target deviation is not zero, then the corresponding initial adjustment factor is found from the preset deviation and factor mapping table corresponding to the target body shape to obtain several initial adjustment factors; the target body shape corresponds one-to-one with the preset deviation and factor mapping table. Based on all the initial adjustment factors, the score adjustment factors are determined.
5. The method of coaching sports based on computer vision and sports biomechanics differences as claimed in claim 4, wherein, The determination of the scoring adjustment factors based on all the initial adjustment factors includes: Calculate the product of all the initial adjustment factors to obtain the score adjustment factor; Accordingly, adjusting the initial action score according to the scoring adjustment factor to obtain the final action score includes: The final action score is obtained by multiplying the scoring adjustment factor by the initial action score.
6. The method of coaching sports based on computer vision and sports biomechanics differences as claimed in claim 4, wherein, The calculation of the target deviation between each target physical parameter and the corresponding target parameter range includes: For each target physical parameter, if the target physical parameter is within the corresponding range of the target parameter, then zero is taken as the target deviation; For each target physical parameter, if the target physical parameter is not within the corresponding range of the target parameters, the difference between each target physical parameter and the target parameter in the corresponding range of the target parameters is calculated, and the difference is used as the target deviation; the target parameter is the range boundary parameter that is closest to the target physical parameter within the range of the standard physical parameters.
7. The exercise guidance method based on computer vision and sports biomechanical differences according to any one of claims 1 to 6, characterized in that, The step of obtaining the sports type corresponding to the target user includes: The exercise mode and exercise goal selected by the target user are obtained, and the exercise type is determined based on the exercise mode and exercise goal; the exercise type includes exercise posture type, dominant muscle group type and energy metabolism type.
8. A motion guidance device based on computer vision and differences in sports biomechanics, characterized in that, include: The information acquisition module is used to acquire the physical parameters of the target user based on the visual camera, and to acquire the corresponding movement type of the target user; The scoring adjustment factor determination module is used to determine the standard physical parameter range of the target user under the exercise type based on the physical parameters, and to determine the scoring adjustment factor based on the deviation between the physical parameters and the standard physical parameter range; The exercise biomechanical information acquisition module is used to determine the target user's optimal exercise biomechanical information under the exercise type by using the exercise biomechanical knowledge base and based on the physical parameters, and to acquire the target user's real movements based on the visual camera and inertial measurement unit sensor to determine the real exercise biomechanical information. The comprehensive scoring module is used to analyze the target user's actual motion biomechanical information based on the target's optimal motion biomechanical information to determine an initial motion score, and adjust the initial motion score according to the scoring adjustment factor to obtain a final motion score. The exercise guidance module is used to provide an exercise guidance report to the target user based on the final movement score, the physical parameters, the standard physical parameter range, the target optimal exercise biomechanical information, and the actual exercise biomechanical information.
9. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor for executing the computer program to implement the motion guidance method based on computer vision and motion biomechanical differences as described in any one of claims 1 to 7.
10. A computer-readable storage medium, characterized in that, Used to store computer programs; wherein, when the computer programs are executed by a processor, they implement the motion guidance method based on computer vision and motion biomechanical differences as described in any one of claims 1 to 7.