A human motion assessment method based on visual images

Abstract

The present invention relates to the technical field of computer vision image processing, and in particular to a human motion assessment method based on visual images, comprising: obtaining a video of a tester to be tested; performing skeleton analysis and posture analysis on a video frame of the video to be tested, and generating a corresponding human key point coordinate diagram; calculating corresponding motion assessment auxiliary information based on the human key point coordinate diagram in combination with corresponding motion assessment standards; completing motion assessment based on the motion assessment auxiliary information in combination with corresponding motion assessment decision basis to generate a corresponding motion assessment result. The human motion assessment method in the present invention can be applied to a variety of motion assessments, thereby improving the efficiency of human motion assessment.

Description

Technical Field

[0001] The present invention relates to the technical field of computer vision image processing, and in particular to a human motion assessment method based on visual images. Background Art

[0002] With the widespread application of Internet big data information technology, the application scenarios of human behavior detection and recognition technology based on visual images are increasing. By analyzing the movement characteristics of human expressions, postures, etc., it can provide rich identification feature information for the detection and prediction of people's behavior in public places or specific activity spaces, which is an important part of human activity big data information.

[0003] For example, in the fields of sports competition and health screening, it is necessary to identify human movements in order to evaluate the standard conditions of the movements. Movement evaluation needs to be based on movement evaluation standards and movement evaluation decision-making basis. In the early days, the commonly used evaluation method was that the evaluator observed the test subject's movements with his eyes, and manually compared the test subject's movements with the standard movement patterns to give scores; at the same time, the evaluator used a handheld camera to extract the video, and then took a screenshot of the video and saved it for reference. The above existing methods not only waste manpower and material resources, but also the evaluation results are not objective and accurate due to subjective reasons during human judgment.

[0004] With the development of computer technology, methods for evaluating human motion based on visual images have emerged in the prior art. For example, Chinese patent publication number CN110941990A discloses a method and device for evaluating human motion based on skeletal key points, which includes: collecting action pictures of the target subject during human motion; extracting the skeletal key point coordinates of the target subject's motion based on the action pictures; inputting the skeletal key point coordinates into a pre-trained evaluation model to evaluate the target subject's motion, wherein the evaluation model evaluates human motion based on the human posture azimuth calculated from the skeletal key point coordinates.

[0005] The human motion evaluation method in the above-mentioned existing scheme calculates the corresponding human posture azimuth based on the skeleton key points, thereby realizing the evaluation of human motion. However, the applicant found that the human posture azimuth, as a kind of motion evaluation auxiliary information, can generally only be applied to the evaluation of a corresponding action, and is difficult to be applied to the evaluation of multiple different actions. Because, when evaluating certain actions, it is also necessary to calculate the distance or position relationship between key points, and even the similarity between video frames, etc. However, the specific calculation of which kind of motion evaluation auxiliary information to complete the motion evaluation is related to the corresponding motion evaluation standard and the motion evaluation decision basis, and there is no general evaluation method that can be applied to multiple motion evaluations in the prior art, so that it is necessary to design a corresponding special evaluation method for each action, resulting in low efficiency of motion evaluation. Therefore, how to design a general motion evaluation method applicable to multiple motion evaluations is a technical problem that needs to be solved urgently. Summary of the invention

[0006] In view of the above-mentioned deficiencies in the prior art, the technical problem to be solved by the present invention is: how to provide a human motion assessment method that can be applicable to a variety of motion assessments, thereby improving the efficiency of human motion assessment.

[0007] In order to solve the above technical problems, the present invention adopts the following technical solutions:

[0008] A method for evaluating human motion based on visual images comprises the following steps:

[0009] S1: Obtain the test video of the tester;

[0010] S2: Perform skeleton analysis and posture analysis on the video frames of the video to be tested, and generate a corresponding human body key point coordinate map;

[0011] S3: Calculate the corresponding action evaluation auxiliary information based on the human body key point coordinate diagram and the corresponding action evaluation standard;

[0012] S4: Complete the action evaluation based on the action evaluation auxiliary information combined with the corresponding action evaluation decision basis to generate a corresponding action evaluation result.

[0013] Preferably, in step S3, the action assessment auxiliary information includes a counterclockwise rotation angle between key points;

[0014] Calculate the counterclockwise rotation angle by the following steps:

[0015] S301: Obtain key point coordinates A, B, C for calculating the counterclockwise rotation angle;

[0016] S302: Calculate the corresponding key point vector based on the key point coordinates A, B, and C and the key point vector

[0017] S303: The key point vector Rotate counterclockwise until it is aligned with the key point vector coincide; then the key point vector Rotate to keypoint vector The angle is taken as the corresponding counterclockwise rotation angle.

[0018] Preferably, in step S3, the action evaluation auxiliary information includes the similarity between the counterclockwise rotation angle to be tested in the video to be tested and the counterclockwise rotation angle of the corresponding template in the template video; wherein the similarity between the counterclockwise rotation angle to be tested and the counterclockwise rotation angle of the corresponding template is calculated based on a dynamic time warping algorithm.

[0019] Preferably, the similarity between the counterclockwise rotation angle to be measured and the counterclockwise rotation angle of the corresponding template is calculated by the following steps:

[0020] S311: Obtain the counterclockwise rotation angle sequence P (p 1 ,p 2 ,…,p n ), and the corresponding template counterclockwise rotation angle sequence Q(q 1 ,q 2 ,…,q m );p i Indicates the counterclockwise rotation angle to be measured corresponding to the i-th frame in the video to be measured; q i Indicates the counterclockwise rotation angle of the template corresponding to the i-th frame in the template video;

[0021] S312: constructing an n×m two-dimensional matrix C based on the counterclockwise rotation angle sequence to be tested and the template counterclockwise rotation angle sequence; C(i,j) represents the Euclidean distance between the i-th counterclockwise rotation angle to be tested in the counterclockwise rotation angle sequence to be tested and the j-th template counterclockwise rotation angle in the template counterclockwise rotation angle sequence;

[0022] S313: In the two-dimensional matrix C, the cumulative distance from the starting position C(0,0) to the end position C(n,m) is calculated, and the corresponding matching path is recorded; then the matching path corresponding to the minimum cumulative distance D is selected as the optimal matching path, and the number of path steps K of the optimal matching path is calculated;

[0023] S314: Calculate the corresponding similarity score based on the minimum cumulative distance D and the number of path steps K.

[0024] Preferably, the cumulative distance is calculated by the following formula:

[0025] d(i,j)=c(i,j)+min{d(i-1,j-1),d(i-1,j),d(i,j-1)};

[0026] The optimal matching path is selected by the following formula:

[0027]

[0028] The similarity score is calculated using the following formula:

[0029]

[0030] In the above formula: d(i,j) represents the cumulative distance from the starting position C(0,0) to the end position C(i,j); c k represents the kth element in the two-dimensional matrix C; s represents the similarity score; h represents the adjustment coefficient, which is set to 0.2.

[0031] Preferably, the types of counterclockwise rotation angles include the angle between the left forearm and the left upper arm, the angle between the left upper arm and the left shoulder, the angle between the left upper arm and the torso, the angle between the torso and the left thigh, the angle between the left thigh and the left calf, the angle between the right upper arm and the right forearm, the angle between the right shoulder and the right upper arm, the angle between the torso and the right upper arm, the angle between the torso and the right thigh, and the angle between the right thigh and the right calf; when calculating the similarity between the counterclockwise rotation angle to be tested and the counterclockwise rotation angle of the template, only the similarity between a certain type of counterclockwise rotation angles can be calculated at a time.

[0032] Preferably, in step S3, when calculating the action evaluation auxiliary information, the recommended key points are selected to participate in the calculation through the following steps:

[0033] S321: Calculate the variance of each counterclockwise rotation angle in the human body key point coordinate diagram;

[0034] S322: Calculating the motion information ratio corresponding to the counterclockwise rotation angle based on the variance of each counterclockwise rotation angle;

[0035] S323: Select the key point corresponding to the counterclockwise rotation angle with the largest motion information ratio as the recommended key point.

[0036] Preferably, the variance of the counterclockwise rotation angle is calculated by the following formula:

[0037]

[0038] The motion information ratio of the counterclockwise rotation angle is calculated by the following formula:

[0039]

[0040] In the above formula: 2 represents the variance of the counterclockwise rotation angle; R represents the counterclockwise rotation angle; u rrepresents the mean value of the counterclockwise rotation angle in the human body key point coordinate diagram; N represents the number of counterclockwise rotation angles in the human body key point coordinate diagram; I n Represents the motion information ratio of the nth counterclockwise rotation angle in the human body key point coordinate diagram; It represents the variance of the nth counterclockwise rotation angle in the human body key point coordinate diagram; e represents a natural constant.

[0041] Preferably, in step S3, the action assessment auxiliary information includes the Euclidean distance between key points;

[0042] By formula Calculate the Euclidean distance between key points;

[0043] In the above formula: d(A,B) represents the key point A(x 1 ,x 2 ,…,x n ) and B(y 1 ,y 2 ,…,y n ) is the Euclidean distance between them.

[0044] Preferably, in step S3, the action assessment auxiliary information includes the positional relationship between key points; the positional relationship between key points includes slope and difference.

[0045] Compared with the prior art, the human body motion assessment method of the present invention has the following beneficial effects:

[0046] 1. The present invention can complete the evaluation of multiple actions according to the action evaluation standards and action evaluation decision basis of the corresponding actions, that is, it provides a human motion evaluation method suitable for multiple action evaluations, so that there is no need to design a corresponding evaluation method for each action, thereby improving the efficiency of human motion evaluation.

[0047] 2. The present invention generates a coordinate diagram of key points of the human body by means of skeleton analysis and posture analysis, and then calculates motion evaluation auxiliary information in combination with corresponding motion evaluation standards, and completes motion evaluation in combination with motion evaluation decision basis, so that the calculation of motion evaluation auxiliary information and its motion evaluation can be associated with the corresponding motion evaluation standards and motion evaluation decision basis, thereby ensuring the calculation accuracy of motion evaluation auxiliary information and the accuracy and effect of motion evaluation. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to make the purpose, technical solution and advantages of the invention more clear, the present invention will be further described in detail below with reference to the accompanying drawings, in which:

[0049] Figure 1 The logical block diagram of the human motion evaluation method;

[0050] Figure 2 Schematic diagram of ten counterclockwise rotation angles on the human body;

[0051] Figure 3 It is a schematic diagram of the limb angle between the right upper arm and the right forearm;

[0052] Figure 4 Schematic diagram of the counterclockwise rotation angle between the right upper arm and the right forearm. DETAILED DESCRIPTION

[0053] The following is a further detailed description through specific implementation methods:

[0054] Example:

[0055] First, the meaning of action evaluation criteria and action evaluation decision basis are explained.

[0056] Action assessment criteria: refers to the items that need to be assessed during action assessment.

[0057] Taking squat as an example, the movement evaluation criteria include: 1) whether the test rod is directly above the head; 2) whether the torso and calf are parallel or perpendicular to the ground; 3) whether the thigh is below the horizontal line when squatting; 4) whether the knees remain in the same direction as the feet.

[0058] Action evaluation decision basis: refers to the scoring criteria for action evaluation.

[0059] Taking squat as an example, the decision-making basis for action evaluation includes: 1) the test rod is directly above the head, the torso is parallel to the calf or perpendicular to the ground, the thighs are below the horizontal line when squatting, and the knees are kept in the same direction as the feet, and 3 points are scored; 2) Failure to complete the required action or placing a wooden board under the heels to complete the required action will score 2 points; 3) Failure to complete the required action even with a template under the heels will score 1 point; 4) Pain in any part of the body during the test will score 0 points.

[0060] Based on the above description, this embodiment discloses a human motion assessment method based on visual images.

[0061] like Figure 1 As shown, the human action evaluation method based on visual images includes the following steps:

[0062] S1: Obtain the test video of the tester;

[0063] S2: Perform skeleton analysis and posture analysis on the video frames of the video to be tested, and generate a corresponding human body key point coordinate map; in this embodiment, the AlphaPose model proposed by Shanghai Jiao Tong University is used to perform skeleton analysis and posture analysis on the video frames of the video to be tested.

[0064] S3: Calculate the corresponding action evaluation auxiliary information based on the human body key point coordinate diagram and the corresponding action evaluation standard;

[0065] S4: Complete the action evaluation based on the action evaluation auxiliary information combined with the corresponding action evaluation decision basis to generate a corresponding action evaluation result.

[0066] In the present invention, the evaluation of the corresponding multiple actions can be completed according to the action evaluation standard and action evaluation decision basis of the corresponding action, that is, a human body action evaluation method suitable for multiple action evaluation is provided, so that there is no need to design a corresponding evaluation method for each action, thereby improving the efficiency of human body action evaluation. At the same time, the present invention generates a human body key point coordinate diagram by means of skeleton analysis and posture analysis, and then calculates action evaluation auxiliary information in combination with the corresponding action evaluation standard, and completes the action evaluation in combination with the action evaluation decision basis, so that the calculation of the action evaluation auxiliary information and its action evaluation can be associated with the corresponding action evaluation standard and action evaluation decision basis, thereby ensuring the calculation accuracy of the action evaluation auxiliary information and the accuracy and effect of the action evaluation.

[0067] In the specific implementation process, the action evaluation auxiliary information includes the counterclockwise rotation angle between key points; combined with Figure 2 As shown, the types of counterclockwise rotation angles include the angle between the left forearm and the left upper arm, the angle between the left upper arm and the left shoulder, the angle between the left upper arm and the torso, the angle between the torso and the left thigh, the angle between the left thigh and the left calf, the angle between the right upper arm and the right forearm, the angle between the right shoulder and the right upper arm, the angle between the torso and the right upper arm, the angle between the torso and the right thigh, and the angle between the right thigh and the right calf.

[0068] Calculate the counterclockwise rotation angle by the following steps:

[0069] S301: Obtain key point coordinates A, B, C for calculating the counterclockwise rotation angle;

[0070] S302: Calculate the corresponding key point vector based on the key point coordinates A, B, and C and the key point vector

[0071] S303: The key point vector Rotate counterclockwise until it is aligned with the key point vector coincide; then the key point vector Rotate to keypoint vector The angle is taken as the corresponding counterclockwise rotation angle.

[0072] The skeleton analysis and posture analysis of the video frame results in a two-dimensional posture. The key points in the human body key point coordinate diagram are actually the projection of the real posture on the two-dimensional plane, which makes it difficult for a simple limb angle to accurately represent the limb characteristics of an action. Figure 3 As shown in the figure, when the right arm is bent in front of the chest and bent at the side of the body, the limb angle between the right upper arm and the right forearm is the same. From the data, the limb characteristics of the right arm movement are the same, because the limb angle between the right upper arm and the right forearm is the same, but in fact, there are great differences between the two movements.

[0073] To this end, the present invention adds direction information, i.e., rotation direction, based on the limb angle, so that the generated counterclockwise rotation angle has both angle information and direction information (e.g. Figure 4 As shown in FIG. 1 , the problem of partial posture information loss when the real posture is projected onto a two-dimensional plane can be solved, and the characteristics of the action limbs can be accurately represented, thereby ensuring the accuracy of human motion evaluation. At the same time, the ten counterclockwise rotation angles designed by the present invention can basically cover the important action limb characteristics of the human posture, thereby further ensuring the effect of human motion evaluation.

[0074] In a specific implementation process, the action assessment auxiliary information includes the similarity between the counterclockwise rotation angle to be tested in the video to be tested and the counterclockwise rotation angle of the corresponding template in the template video; wherein the similarity between the counterclockwise rotation angle to be tested and the counterclockwise rotation angle of the corresponding template is calculated based on the dynamic time warping algorithm. In this embodiment, when calculating the similarity between the counterclockwise rotation angle to be tested and the counterclockwise rotation angle of the template, only the similarity between a certain type of counterclockwise rotation angle can be calculated at a time.

[0075] The similarity between the counterclockwise rotation angle to be tested and the counterclockwise rotation angle of the corresponding template is calculated by the following steps:

[0076] S311: Obtain the counterclockwise rotation angle sequence P (p 1 ,p 2 ,…,p n ), and the corresponding template counterclockwise rotation angle sequence Q(q 1 ,q 2 ,…,q m );p i Indicates the counterclockwise rotation angle to be measured corresponding to the i-th frame in the video to be measured; q i Indicates the counterclockwise rotation angle of the template corresponding to the i-th frame in the template video;

[0077] S312: constructing an n×m two-dimensional matrix C based on the counterclockwise rotation angle sequence to be tested and the template counterclockwise rotation angle sequence; C(i,j) represents the Euclidean distance between the i-th counterclockwise rotation angle to be tested in the counterclockwise rotation angle sequence to be tested and the j-th template counterclockwise rotation angle in the template counterclockwise rotation angle sequence;

[0078] S313: In the two-dimensional matrix C, the cumulative distance from the starting position C(0,0) to the end position C(n,m) is calculated, and the corresponding matching path is recorded; then the matching path corresponding to the minimum cumulative distance D is selected as the optimal matching path, and the number of path steps K of the optimal matching path is calculated;

[0079] S314: Calculate the corresponding similarity score based on the minimum cumulative distance D and the number of path steps K.

[0080] In the specific implementation process, the cumulative distance is calculated by the following formula:

[0081] d(i,j)=c(i,j)+min{d(i-1,j-1),d(i-1,j),d(i,j-1)};

[0082] The optimal matching path is selected by the following formula:

[0083]

[0084] The similarity score is calculated using the following formula:

[0085]

[0086] In the above formula: d(i,j) represents the cumulative distance from the starting position C(0,0) to the end position C(i,j); c k represents the kth element in the two-dimensional matrix C; s represents the similarity score; h represents the adjustment coefficient, which is set to 0.2. The DTW in the formula refers to the optimal matching path algorithm.

[0087] When evaluating actual actions, it is necessary to compare the video to be tested with the template video and calculate the similarity, and then complete the action evaluation through the similarity. Video images generally exist in the form of time series, so what needs to be calculated is the similarity between the two time series. However, since different people perform the same action at different speeds, and even the same person repeats the same action, there will be differences between body parts, resulting in the lengths of the two time series being basically inconsistent. At this point, the existing method of calculating similarity based on Euclidean distance cannot effectively calculate the similarity between time series.

[0088] To this end, the present invention introduces a "dynamic time warping algorithm" through the above steps to calculate the similarity between the video to be tested and the template video, and calculates the cumulative minimum distance between the two by adjusting the time correspondence (i.e., length relationship) of the two sequences to find the optimal matching path, so that the similarity of the time series of two sequences of unequal lengths can be calculated, thereby ensuring the accuracy of human motion evaluation. At the same time, since the counterclockwise rotation angle can accurately represent the characteristics of the body parts of the action, the present invention can effectively represent the similarity between the video to be tested and the template video through the similarity between the counterclockwise rotation angles, thereby better assisting in completing the human motion evaluation.

[0089] In the specific implementation process, when calculating the auxiliary information of action evaluation, the recommended key points are selected to participate in the calculation through the following steps:

[0090] S321: Calculate the variance of each counterclockwise rotation angle in the human body key point coordinate diagram;

[0091] S322: Calculating the motion information ratio corresponding to the counterclockwise rotation angle based on the variance of each counterclockwise rotation angle;

[0092] S323: Select the key point corresponding to the counterclockwise rotation angle with the largest motion information ratio as the recommended key point.

[0093] The variance of the counterclockwise rotation angle is calculated using the following formula:

[0094]

[0095] The motion information ratio of the counterclockwise rotation angle is calculated by the following formula:

[0096]

[0097] In the above formula: 2 represents the variance of the counterclockwise rotation angle; R represents the counterclockwise rotation angle; u r represents the mean value of the counterclockwise rotation angle in the human body key point coordinate diagram; N represents the number of counterclockwise rotation angles in the human body key point coordinate diagram; I n Represents the motion information ratio of the nth counterclockwise rotation angle in the human body key point coordinate diagram; It represents the variance of the nth counterclockwise rotation angle in the human body key point coordinate diagram; e represents a natural constant.

[0098] In actual movement assessment, in most movements, only a few limbs perform the main movement, while other limbs do not move much or at all. Among them, the angle change range of the limbs that perform the main movement is larger, while the angle change range of the limbs that do not perform the movement is smaller. When performing movement assessment, generally only the limbs that perform the main movement can be considered.

[0099] To this end, the present invention selects recommended key points with large motion amplitude to participate in the calculation by calculating the variance of the counterclockwise rotation angle, the motion information ratio and the Euclidean distance difference between key points. On the one hand, the recommended key points can accurately reflect the limbs performing the main movement, thereby ensuring the accuracy of human motion evaluation; on the other hand, the key points corresponding to the limbs with small motion amplitude or no motion are not involved in the calculation, which can reduce the calculation amount of human motion evaluation.

[0100] In the specific implementation, the auxiliary information of action evaluation includes the Euclidean distance between key points;

[0101] By formula Calculate the Euclidean distance between key points;

[0102] In the above formula: d(A,B) represents the key point A(x 1 ,x 2 ,…,x n ) and B(y 1 ,y 2 ,…,y n ) is the Euclidean distance between them.

[0103] During the specific implementation process, the action evaluation auxiliary information includes the positional relationship between key points; the positional relationship between key points includes slope and difference.

[0104] In the present invention, it is also possible to calculate action evaluation auxiliary information such as the Euclidean distance between key points and the positional relationship between key points to assist in completing action evaluation, so that it can be better applied to the evaluation of various actions, thereby ensuring the efficiency and accuracy of human body action evaluation.

[0105] Specifically, when generating a human body key point coordinate map, firstly, the corresponding video frame is input into a pre-trained posture estimation model, and the corresponding heat map is output; then, the key point coordinates are calculated by the heat map to obtain the corresponding human body key point coordinate map;

[0106] When training the posture estimation model, a posture data set for training is obtained; then the labels pre-marked on the training graph in the posture data set are converted into corresponding heat map labels to obtain corresponding label heat maps; finally, the posture estimation model is trained based on the label heat maps;

[0107] When generating a label heat map, first set the size of the label heat map (W h ×H h ) to generate a size W h ×H h The all-zero matrix is ​​then used to calculate the heat distribution of the pre-labeled labels on the label heat map to generate the corresponding label heat map;

[0108]

[0109] When calculating the coordinates of the key points, get the size W h ×H h The heat map is then reduced to 1*W h *H h ; Then calculate the maximum heat value index of the corresponding key point in the heat map by the following formula; Finally, calculate the coordinates of the corresponding key point by combining the index corresponding to the maximum heat value in the heat map with the size of the heat map. Specifically, divide the index by the corresponding W h , the quotient is of size W h ×H h The number of rows of key points is x, and the remainder is of size W h ×H h The number of columns of the key points y, that is, the coordinates of the key points (x, y);

[0110]

[0111] In the above formula: g represents the heat value; x 0 ,y 0 represents the real coordinates of the pre-annotated label; x, y represent the coordinates of the label in the label heat map; σ represents the standard deviation, which takes a value of 2 or 3; e represents a natural constant; i, j represent the index of the one-dimensional heat map respectively; x i 、x j represents the heat value corresponding to index i and j; β represents the calibration coefficient.

[0112] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit them. Although the present invention has been described with reference to the preferred embodiments of the present invention, it should be understood by those skilled in the art that various changes can be made to it in form and detail without departing from the spirit and scope of the present invention as defined in the appended claims. At the same time, common sense such as the well-known specific structures and characteristics in the embodiments are not described in detail here. Finally, the scope of protection claimed by the present invention shall be based on the content of its claims, and the specific implementation methods and other records in the specification can be used to interpret the content of the claims.

Claims

1. A method for evaluating human motion based on visual images, characterized in that: The following steps are involved: S1: Obtain the test video of the tester; S2: Perform skeleton analysis and posture analysis on the video frames of the video to be tested, and generate a corresponding human body key point coordinate map; S3: Calculate the corresponding action evaluation auxiliary information based on the human body key point coordinate diagram and the corresponding action evaluation standard; In step S3, the action evaluation auxiliary information includes the counterclockwise rotation angle between the key points; Calculate the counterclockwise rotation angle by the following steps: S301: Obtain key point coordinates A, B, C for calculating the counterclockwise rotation angle; S302: Calculate the corresponding key point vector based on the key point coordinates A, B, and C and the key point vector S303: The key point vector Rotate counterclockwise until it is aligned with the key point vector coincide; then the key point vector Rotate to keypoint vector The angle of is taken as the corresponding counterclockwise rotation angle; The action assessment auxiliary information includes the similarity between the counterclockwise rotation angle to be tested in the video to be tested and the counterclockwise rotation angle of the corresponding template in the template video; wherein the similarity between the counterclockwise rotation angle to be tested and the counterclockwise rotation angle of the corresponding template is calculated based on a dynamic time warping algorithm; When calculating the auxiliary information for action evaluation, the recommended key points are selected for calculation through the following steps: S321: Calculate the variance of each counterclockwise rotation angle in the human body key point coordinate diagram; S322: Calculating the motion information ratio corresponding to the counterclockwise rotation angle based on the variance of each counterclockwise rotation angle; S323: Selecting the key point corresponding to the counterclockwise rotation angle with the largest motion information ratio as the recommended key point; The variance of the counterclockwise rotation angle is calculated using the following formula: The motion information ratio of the counterclockwise rotation angle is calculated by the following formula: In the above formula: 2 represents the variance of the counterclockwise rotation angle; R represents the counterclockwise rotation angle; u r represents the mean value of the counterclockwise rotation angle in the human body key point coordinate diagram; N represents the number of counterclockwise rotation angles in the human body key point coordinate diagram; I n Represents the motion information ratio of the nth counterclockwise rotation angle in the human body key point coordinate diagram; represents the variance of the nth counterclockwise rotation angle in the human body key point coordinate diagram; e represents a natural constant; S4: Complete the action evaluation based on the action evaluation auxiliary information combined with the corresponding action evaluation decision basis to generate a corresponding action evaluation result.

2. The method for evaluating human motion based on visual images as claimed in claim 1, characterized in that: The similarity between the counterclockwise rotation angle to be tested and the counterclockwise rotation angle of the corresponding template is calculated by the following steps: S311: Obtain the counterclockwise rotation angle sequence P (p1, p2, ..., p n ), and the corresponding template counterclockwise rotation angle sequence Q(q1,q2,…,q m );p i Indicates the counterclockwise rotation angle to be measured corresponding to the i-th frame in the video to be measured; q i Indicates the counterclockwise rotation angle of the template corresponding to the i-th frame in the template video; S312: constructing an n×m two-dimensional matrix C based on the counterclockwise rotation angle sequence to be tested and the template counterclockwise rotation angle sequence; C(i,j) represents the Euclidean distance between the i-th counterclockwise rotation angle to be tested in the counterclockwise rotation angle sequence to be tested and the j-th template counterclockwise rotation angle in the template counterclockwise rotation angle sequence; S313: In the two-dimensional matrix C, the cumulative distance from the starting position C(0,0) to the end position C(n,m) is calculated, and the corresponding matching path is recorded; then the matching path corresponding to the minimum cumulative distance D is selected as the optimal matching path, and the number of path steps K of the optimal matching path is calculated; S314: Calculate the corresponding similarity score based on the minimum cumulative distance D and the number of path steps K.

3. The method for assessing human motion based on visual images as claimed in claim 2, characterized in that: The cumulative distance is calculated using the following formula: d(i,j)=c(i,j)+min{d(i-1,j-1),d(i-1,j),d(i,j-1)}; The optimal matching path is selected by the following formula: The similarity score is calculated using the following formula: In the above formula: d(i,j) represents the cumulative distance from the starting position C(0,0) to the end position C(i,j); c k represents the kth element in the two-dimensional matrix C; s represents the similarity score; h represents the adjustment coefficient, which is set to 0.

2.

4. The method for assessing human motion based on visual images as claimed in claim 1, characterized in that: The types of counterclockwise rotation angles include the angle between the left forearm and the left upper arm, the angle between the left upper arm and the left shoulder, the angle between the left upper arm and the trunk, the angle between the trunk and the left thigh, the angle between the left thigh and the left calf, the angle between the right upper arm and the right forearm, the angle between the right shoulder and the right upper arm, the angle between the trunk and the right upper arm, the angle between the trunk and the right thigh, and the angle between the right thigh and the right calf; When calculating the similarity between the counterclockwise rotation angle to be tested and the counterclockwise rotation angle of the template, only the similarity between a certain type of counterclockwise rotation angles can be calculated at a time.

5. The method for assessing human motion based on visual images as claimed in claim 1, characterized in that: In step S3, the action evaluation auxiliary information includes the Euclidean distance between key points; By formula Calculate the Euclidean distance between key points; In the above formula: d(A,B) represents the key point A)x1,x2,…,x n ) and B(y1,y2,…,y n ) is the Euclidean distance between them.

6. The method for assessing human motion based on visual images as claimed in claim 1, characterized in that: In step S3, the action assessment auxiliary information includes the positional relationship between key points; the positional relationship between key points includes slope and difference.

Images