A scoring system, method and apparatus for freestyle scooter competitions

Abstract

The application relates to the technical field of computer application, and discloses a scoring system, method and equipment for a freestyle scooter competition, wherein the system comprises an image acquisition module, which is used for acquiring a video stream of a competition of a target object, analyzing one or more action units of the target object in the video stream, and acquiring a plurality of video frames in the action units; an image analysis module, which is used for determining a plurality of action parameters in the action units based on the plurality of video frames of each action unit; an upper computer module, which is used for inputting the action units into a scoring architecture model, evaluating each action unit according to the plurality of action parameters of each action unit, and obtaining a comprehensive score of the target object based on an evaluation result; and a display module, which is used for acquiring the comprehensive score of the target object and displaying the comprehensive score to a terminal screen. The technical scheme provided by the application can quickly and accurately obtain a competition score of a freestyle scooter player in a complex environment.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of computer application, and in particular to a scoring system, method and device for a freestyle scooter competition. BACKGROUND

[0002] With the continuous development of the computer industry, big data models are increasingly popular in various fields. In the field of competitive sports, freestyle scooter is a new sport that was officially listed as an Olympic competition event in 2020.

[0003] However, freestyle scooter competition requires athletes to use various auxiliary terrain to complete a series of aerial tricks, and the complexity of the movements makes it difficult for judges to score. With the continuous development of freestyle scooter and the continuous improvement of athletes, there is an urgent need for a solution that can quickly and accurately determine the competition score in a complex environment. SUMMARY

[0004] The present application provides a scoring system, method and device for a freestyle scooter competition, which can quickly and accurately determine the competition results of freestyle scooter athletes.

[0005] The first aspect of the present application provides a scoring system for a freestyle scooter competition, the system comprising an image acquisition module, an image analysis module, a host computer module, and a display module, wherein the host computer module runs a scoring architecture model, wherein: the image acquisition module is configured to obtain a video stream of a target object's competition, analyze one or more action units of the target object in the video stream, and obtain a plurality of video frames in the action unit; the image analysis module is configured to determine a plurality of action parameters in each action unit based on each action unit and a plurality of video frames in the action unit; the host computer module is configured to obtain the one or more action units and input the action units into a pre-set scoring architecture model, wherein the scoring architecture model evaluates each action unit based on a plurality of action parameters of each action unit, and obtains a comprehensive score of the target object based on the evaluation result; and the display module is configured to obtain the comprehensive score of the target object in the host computer module and display the comprehensive score on a terminal screen.

[0006] In one possible implementation, the image acquisition module comprises an image acquisition unit and an image processing unit, wherein: the image acquisition unit is configured to obtain a video stream of a target object's competition; and the image processing unit is configured to divide one or more action units of the target object based on the video stream, and obtain a plurality of video frames in an action unit, wherein the action unit is the action process of the target object performing one aerial movement.

[0007] In a possible implementation, the image analysis module comprises a height detection unit, a time detection unit, a motion continuity detection unit and a motion completeness detection unit, and the motion parameters comprise a take-off height, a hang time, a motion continuity parameter and a motion completeness parameter, wherein: the height detection unit is configured to acquire first motion data of a current motion unit, and determine a take-off height of the target object according to the first motion data; the time detection unit is configured to acquire second motion data of the current motion unit and time stamps corresponding to each video frame, and determine a hang time of the target object according to the second motion data and the time stamps corresponding to each video frame; the motion continuity detection unit is configured to acquire a plurality of video frames of the target object in one motion unit and time stamps of the video frames, and determine a motion continuity parameter of the target object according to the video frames and the time stamps; and the motion completeness detection unit is configured to acquire a standard motion module, the standard motion module comprising a plurality of specified standard motions, compare each motion unit with a corresponding specified standard motion, and determine a motion completeness parameter based on a comparison result.

[0008] In a possible implementation, the host computer module comprises a basic scoring unit and a comprehensive scoring unit, and a scoring architecture model runs in the basic scoring unit, wherein: the basic scoring unit is configured to acquire each motion unit of the target object, input a video stream of each motion unit into the scoring architecture model, and perform evaluation on each motion unit according to a preset motion standard by using the scoring architecture model, obtain a standard motion code of each motion unit based on an evaluation result, and determine a standard score of each motion unit according to the standard motion code of each motion unit; and the comprehensive scoring unit is configured to acquire the standard scores of the motion units, perform weighted calculation on a standard score corresponding to a current motion unit according to a motion parameter of each motion unit, obtain a motion score of each motion unit based on a weighted calculation result, and obtain a comprehensive score of the target object based on the motion scores of the motion units.

[0009] The second aspect of the application provides a scoring method for a freestyle scooter competition, the method comprising: acquiring a video stream of a competition of a target object by an image acquisition module, and parsing one or more action units of the target object in the video stream to acquire a plurality of video frames in each action unit; determining a plurality of action parameters of the target object in each action unit based on the action units and the plurality of video frames of the action units by an image analysis module, and transmitting the action parameters to an upper computer module; inputting the plurality of action parameters to a preset scoring architecture model by the upper computer module, evaluating each action unit according to the plurality of action parameters of each action unit by the scoring architecture model, obtaining a comprehensive score of the target object based on the evaluation result, and transmitting the comprehensive score to a display module; and receiving the comprehensive score by the display module and displaying the comprehensive score on a terminal screen.

[0010] The third aspect of the application provides a computer device, the device comprising a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, and the processor executing the computer instructions to perform the scoring method for a freestyle scooter competition.

[0011] The technical solution provided by the embodiment of the application can improve the scoring speed and accuracy of the freestyle scooter competition by acquiring a video stream of a competition of a target object and automatically scoring the competition process of the target object. In the embodiment, a player is selected as the target object, a video stream of the competition of the player is acquired by an image acquisition module, and the video stream is divided into different action units by an image analysis module. An upper computer module analyzes and evaluates each action unit to obtain a plurality of action parameters of each action unit and a standard score of each action unit, and then calculates a comprehensive score of the player in the video stream, which is taken as the competition result of the player.

[0012] It can be seen that the technical solution provided by the embodiment of the application can quickly and accurately obtain the competition result of a freestyle scooter player. BRIEF DESCRIPTION OF DRAWINGS

[0013] In order to more clearly illustrate the technical solutions in the embodiments of the application or the prior art, the drawings needed in the description of the embodiments or the prior art will be briefly introduced. Obviously, the drawings in the following description are some embodiments of the application, and other drawings can be obtained by those skilled in the art without creative effort.

[0014] Figure 1A structural schematic diagram of a scoring system of a free-style scooter race is provided for an embodiment of the present application.

[0015] Figure 2 A step schematic diagram of a scoring method of a free-style scooter race is provided for an embodiment of the present application.

[0016] Figure 3 A structural schematic diagram of a computer device is provided for an embodiment of the present application. DETAILED DESCRIPTION

[0017] In order to make the objectives, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some but not all of the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0018] In addition, the descriptions involving “first”, “second” and the like in the present application are only for the purpose of description, and cannot be understood as indicating or implying the relative importance of the indicated technical features or implicitly indicating the number of the indicated technical features. Therefore, the features limited by “first”, “second” can explicitly or implicitly include at least one of the features. In the description of the embodiments of the present application, unless otherwise specified, the meaning of “multiple” is two or more. In addition, the use of “based on” or “according to” means openness and inclusiveness, because the process, step, calculation or other action “based on” or “according to” one or more stated conditions or values can be based on additional conditions or values beyond the stated values in practice.

[0019] With the rapid development of the computer industry, big data models are increasingly widely used in various fields. As a new type of challenging sport, freestyle scooter has received widespread attention since it became an official Olympic competition in 2020. Freestyle scooter athletes need to use various auxiliary terrain in the competition to complete a series of high-difficulty aerial trick movements. However, due to the complexity of the terrain and the diversity of the movements, it is easy to cause certain scoring errors and too long consideration time for the judges due to the obstruction of the view and the difficulty in seeing the details of the movements, so the judges often face great challenges in the scoring process. In some application scenarios, there is an urgent need for a new solution that can quickly and accurately give a fair score for the athletes' performance in the competition in a complex competition environment.

[0020] Therefore, one or more embodiments of the present application provide a scoring system, method and device for a freestyle scooter competition, which can solve the above problems, and quickly and accurately obtain the competition score of a freestyle scooter player by fully considering the action difficulty and completion degree of the player.

[0021] Referring to Figure 1 In one embodiment of the present application, a scoring system for a freestyle scooter competition is provided, which comprises an image acquisition module, an image analysis module, a host computer module and a display module. The host computer module runs a scoring architecture model, wherein:

[0022] The image acquisition module is configured to acquire a video stream of a competition of a target object, analyze one or more action units of the target object in the video stream, and acquire a plurality of video frames in the action units.

[0023] The image analysis module is configured to determine a plurality of action parameters in each action unit based on each action unit and the plurality of video frames in the action unit.

[0024] The host computer module is configured to acquire the one or more action units, input the action units into a preset scoring architecture model, and evaluate each action unit according to the plurality of action parameters of each action unit to obtain a comprehensive score of the target object based on the evaluation result.

[0025] The display module is configured to acquire the comprehensive score of the target object in the host computer module, and display the comprehensive score on a terminal screen.

[0026] In the embodiment, the image acquisition module comprises an image acquisition unit and an image processing unit. The image acquisition unit is configured to acquire a video stream of a competition of a target object. The image processing unit is configured to divide one or more action units of the target object according to the video stream, and acquire a plurality of video frames in one action unit, wherein one action unit is an action process of one jump of the target object.

[0027] Specifically, the image acquisition unit acquires a video stream of a target object. For example, a common webcam or a depth camera such as Kinect, Intel Realsense, etc. can be used to directly capture a video stream of a player in a competition, or a live video stream can be acquired through an API interface of a network protocol or a network platform.

[0028] Further, after the image acquisition unit acquires the video stream of the competition, the image processing unit extracts one or more action units of the target object in the video stream. In the freestyle scooter competition, the athlete is the target object to be detected, and the process of performing one jump action is an action unit. The target object in the video frame can be detected by a target detection algorithm, the position of the target object in the video stream can be tracked by a target tracking algorithm, once the start and end of the jump action are identified, an action unit is identified, the start frame and the end frame of each action unit are determined, and a plurality of video frames are extracted from an action unit.

[0029] In the embodiment, the image analysis module includes a height detection unit, a time detection unit, an action continuity detection unit, and an action completeness detection unit, and the action parameters include the jump height, the air time, the action continuity parameter, and the action completeness parameter, wherein:

[0030] The height detection unit is configured to acquire first motion data of a current action unit, and determine the jump height of the target object according to the first motion data.

[0031] The time detection unit is configured to acquire second motion data of the current action unit and time stamps corresponding to each video frame, and determine the air time of the target object according to the second motion data and the time stamps corresponding to each video frame.

[0032] The action continuity detection unit is configured to acquire a plurality of video frames of the target object in an action unit and time stamps of the video frames, and determine the action continuity parameter of the target object according to the video frames and the time stamps.

[0033] The action completeness detection unit is configured to acquire a standard action module, the standard action module includes a plurality of specified standard actions, compare each action unit with the corresponding specified standard action, and determine the action completeness parameter based on the comparison result.

[0034] In an embodiment, the first motion data includes but is not limited to the position coordinates of the target object. The jump height of the target object can be determined by identifying the position coordinates of the target object. In an action unit, a unified reference baseline is first determined, and the ground can be used as the reference baseline. The center point coordinates of the target object in each video frame are acquired by using a target detection algorithm, the vertical distance between the center point of the target object and the reference baseline is calculated, and the distance is taken as the relative height of the current video frame. The relative heights of each video frame are compared, and the relative height value farthest from the reference baseline is taken as the jump height.

[0035] In one embodiment, the second motion data includes, but is not limited to, the vertical position coordinate, velocity and acceleration of the target object, and the duration of the target object in the air is determined by identifying the position coordinate, velocity and acceleration of the target object. The target object is continuously tracked using a target detection algorithm and a target tracking algorithm to determine the takeoff time and the landing time of the target object.

[0036] Alternatively, the takeoff time and the landing time are determined according to the velocity and acceleration of the target object. When the vertical acceleration of the target object changes from 0 m / s 2 to -9.8 m / s 2 , and the vertical velocity of the target object is upward, it is determined that the time is the takeoff time of the target object; when the vertical acceleration of the target object changes from -9.8 m / s 2 to 0 m / s 2 , and the vertical velocity of the target object is downward, it is determined that the time is the landing time of the target object. Further, the video frames of the takeoff time and the landing time of the target object are obtained, and the time stamps of the video frames of the takeoff time and the landing time are extracted, and the difference between the time stamps of the two times is the duration of the target object in the air.

[0037] Alternatively, the takeoff time and the landing time are determined according to the vertical position coordinate of the target object in each frame of a motion unit. The vertical position coordinate of the wheel of the small wheeled vehicle in each video frame is extracted, and the difference between the vertical position coordinates of the wheel between adjacent video frames is calculated. When the difference changes from close to 0 to positive and the trend is continuously increasing, it is determined that the current time is the takeoff time; when the difference changes from positive to close to 0 and then no longer changes, it is determined that the current time is the landing time. Further, the video frames of the takeoff time and the landing time of the target object are obtained, and the time stamps of the video frames of the takeoff time and the landing time are extracted, and the duration of the target object in the air can be calculated according to the time stamps of the two times.

[0038] In one embodiment, the motion labels are aggregated into a motion sequence according to the time stamps of the plurality of video frames to determine the motion continuity parameter. First, all video frames with time stamps in a motion unit are obtained, and the motion labels of each video frame are identified in these video frames. The motion label can be the spatial distribution of the specified part of the target object. All video frames are arranged in ascending order according to the time stamps to ensure that the aggregation of the motion labels is in the time order of the video stream, forming a motion sequence. The motion sequence is input into a continuity detection model for detection, and the continuity parameter is obtained based on the detection result.

[0039] Specifically, the continuity detection model can be trained according to the action sequence samples. A series of action sequence samples are collected, and the action sequence samples cover action sequences of various action types and various continuity parameters. Each action sequence sample is labeled by an expert or a trained labeler, and a continuity label is added. A suitable machine learning model such as a support vector machine model or a neural network model is selected for sample training to form a continuity detection model. The labeled action sequence samples are used to train the continuity detection model, so that the continuity of the action sequence can be predicted. The trained continuity model is evaluated, and cross-validation can be used to evaluate the performance of the model. According to the evaluation result, the model parameters are adjusted to improve the accuracy of continuity recognition. Subsequently, if continuity detection is required, the action sequence to be detected is input into the continuity detection model, and the continuity parameter of the action sequence can be obtained.

[0040] In an optional embodiment, the action sequence samples can also be labeled using One-hot encoding. Each action sequence sample corresponds to a One-hot encoded label, which has a form such as [0, 0, 1, 0, …, 0], that is, a unique label is assigned to each action sequence sample to predict its most likely classification. In TensorFlow (TensorFlow, a data flow programming-based symbolic mathematical system), tf.keras (tf.keras, a TensorFlow API interface) can be used to build a continuity detection model, and the model.fit function can be used to train the continuity detection model. The continuity detection model learns according to the preset training rounds, and continuously iterates the labeled training data to optimize itself, thereby improving the accuracy of the model prediction.

[0041] In an embodiment, the action completeness parameter of each action is determined by comparing the action identifier of each video frame with the specified standard action. First, a plurality of standard action modules are obtained, and each standard action module includes a plurality of specified standard actions arranged in time sequence. Then, the action identifier of each video frame in the action unit is obtained, and the action identifier is compared with the corresponding specified standard action in multi-dimensional feature matching. The first action feature group of the actual action unit and the second action feature group of the specified standard action can be compared, and the completeness of each action is determined based on the comparison result, wherein the first action feature group contains one or more first action features, and the second action feature group contains one or more second action features. The first action feature and the second action feature type include but are not limited to posture feature and amplitude feature.

[0042] Specifically, the standard action module can be collected according to specific scenarios. In the field of freestyle scooter, professional personnel need to define a plurality of representative standard action modules according to possible competition actions, and describe a plurality of specified standard actions in each standard action module in detail, including but not limited to action posture and action amplitude. According to the actual action sequence of the freestyle scooter player, all the specified standard actions in each standard action module are arranged in time sequence.

[0043] Specifically, the action identification of each video frame in an action unit is matched with the corresponding specified standard action in multiple dimensions. The above-mentioned feature matching can include posture matching, amplitude matching, angle matching, etc. According to the results of multi-dimensional matching, the action completeness parameter is calculated by using weighted average method, for example, the posture matching accounts for 60%, the amplitude matching accounts for 40%, the posture matching is 80%, and the amplitude matching is 70%, and the action completeness parameter is 0.76 at this time.

[0044] Exemplarily, in the posture matching, the posture features of the player's competition action are extracted as the first action features, and the posture features of the specified standard action are extracted as the second action features. The above-mentioned posture features include the relative positions of the posture key points of the player's body. The distance between the first action features and the second action features in the posture feature space, such as Euclidean distance, is calculated. The matching degree of the posture features in each distance range is set, and the smaller the distance is, the higher the matching degree is.

[0045] Exemplarily, in the amplitude matching, the amplitude features of the action of the player's competition action are extracted as the first action features, and the amplitude features of the action of the specified standard action are extracted as the second action features. The above-mentioned amplitude features include the amplitude of the action of the player, for example, the distance between the hands and the handlebar is an action amplitude. The amplitude difference between the first action features and the second action features is calculated. The matching degree of the amplitude features in each difference range is set, and the smaller the distance is, the higher the matching degree is.

[0046] In the embodiment, the host computer module includes a basic scoring unit and a comprehensive scoring unit, and the scoring architecture model runs in the basic scoring unit, wherein:

[0047] The basic scoring unit is configured to obtain each action unit of a target object, input a video stream of each action unit into the scoring architecture model, and evaluate each action unit according to a preset action standard. The scoring architecture model obtains a standard action code of each action unit based on the evaluation result, and determines a standard score of each action unit according to the standard action code.

[0048] The comprehensive score unit is configured to obtain a standard score of each action unit, perform weighted calculation on the standard score corresponding to the current action unit according to the action parameter of each action unit, obtain an action score of each action unit based on the weighted calculation result, and obtain a comprehensive score of the target object based on the action scores of the action units.

[0049] In the embodiment, the standard action code of each action unit is obtained first. The video stream of each action unit is input into the scoring architecture model, the scoring architecture model performs preprocessing and feature extraction on the video stream of each action unit to obtain feature data of each action unit, and the standard action code of each action unit is determined according to the feature data and the preset action standard.

[0050] In an optional embodiment, the video stream of each action unit is preprocessed and feature-extracted. Since the color of the image does not affect the action and score of the athlete in the freestyle scooter competition, the above-mentioned action sequence sample can adopt a single-channel grayscale image to reduce the complexity of data.

[0051] Specifically, the scoring architecture model can be trained by using training samples, and the training samples are multiple video samples. TensorFlow is used to load the training samples, so that the training samples are converted from video form to feature form. The training samples are converted into a three-dimensional data matrix, and the dimensions of the matrix are pixel height, width and total frame number respectively. A 3x3x3 convolution kernel is used for convolution operation with a step of 1, and a ReLu (Rectified Linear Unit) activation function is introduced to increase the nonlinearity of the conversion process. Then, 2x2x2 maximum pooling is applied, the largest value in each 2x2x2 region is selected as the representative value of the region, the most important features of the data are retained, the calculation amount in the subsequent training process is effectively reduced, and the training time is shortened. Subsequently, the video data can be converted into a feature map by performing twice convolution and pooling operations, the feature data of each action unit is generated from the feature map through a full connection layer, and the feature classification of different scoring groups is performed according to the feature data, and the scoring groups include action groups, separation of the athlete and the scooter, the number of athlete direction changes and the number of scooter rotation. Among them, the action standards of different scoring groups are different classification features, and the standard branches are different forms of classification features.

[0052] Optionally, the reliability of the scoring architecture model can be evaluated using test samples, which are one or more video samples. Specifically, the test samples are divided into multiple time windows, each window containing enough information to identify an action, the scoring architecture model is used to predict the action in each time window, and the np.argmax function is used to obtain the class index with the highest probability. Arrange the prediction results of all time windows to get the action sequence in the entire video sample. Correspond the predicted action sequence to the scoring rules and give the final result.

[0053] In another optional implementation, feature extraction can extract features related to action units by extracting key frames to reduce data volume and retain key information of actions, for example, extracting a frame every five frames, or extracting a key frame when the position or posture of the action subject between adjacent frames changes by more than a certain threshold. Then, relevant features in the key frames are extracted and classified into different scoring categories, including action categories and action features, including human-vehicle separation, athlete turning frequency, and scooter rotation number. Among them, the above action features are further representations of specific actions in the action category, different scoring categories have different classification features, the category label is the expression form of the classification features of different action categories, and the feature label is the expression form of the classification features of different action features.

[0054] For example, in a freestyle scooter competition, the category label and specific form of the preset action category can be as shown in Table 1, and the category label and corresponding specific form of the preset action feature can be as shown in Tables 2, 3, and 4.

[0055] Table 1 Category label and corresponding action standard of action category

[0056]

[0057] Table 2 Feature label and corresponding specific form of human-vehicle separation

[0058]

[0059] Table 3 Feature label and corresponding specific form of athlete turning frequency

[0060]

[0061] Table 4 Feature label and corresponding specific form of scooter rotation number

[0062]

[0063] In the determination of the action group, the action characteristics of the athlete are determined based on the spatial coordinate axis. According to the change of the body of the athlete relative to the spatial coordinate axis, the X axis is the forward direction of the movement, the Y axis is the horizontal axis perpendicular to the forward direction of the movement, and the Z axis is the vertical axis. Therefore, "no flip" means that the change of the body of the athlete relative to the X, Y and Z axes is less than 180°, "front somersault" means that the body of the athlete rotates more than 180° clockwise around the Y axis, "back somersault" means that the body of the athlete rotates more than 180° counterclockwise around the Y axis, "whirl" means that the body of the athlete rotates more than 180° around the Z axis and the rotation is not divided into left and right, and "spiral" means that the body of the athlete rotates more than 180° around the X axis and the rotation is not divided into left and right.

[0064] Specifically, the above-mentioned standard action code is output in the form of "number + additional character". The number includes four digits, and the four digits represent the action group, the separation of the athlete and the vehicle, the number of changes in direction of the athlete, and the number of rotations of the scooter in order from front to back. The additional character is used to represent some irregular special actions, such as the change of direction action, i.e. the athlete controls the handlebar to rotate in one direction first, and then suddenly changes to rotate in the other direction. This special action can be represented by a special character, for example: H - defined as a change of direction identifier, i.e. change of rotation direction. Other characters can also be used to represent other special actions.

[0065] For example, in a possible action unit, the athlete straightens forward and rotates 360° around the Z axis horizontally after landing, and completes a single-hand rotation of the handlebar twice during the flight, and has turned during the rotation of the handlebar. The standard action code obtained after evaluation of the action unit is "3124H".

[0066] For example, in a possible action unit, the athlete flips 180° backward around the Y axis after landing, and separates the feet from the vehicle during the flight, and the vehicle body rotates one revolution. The standard action code obtained after evaluation of the action unit is "2412".

[0067] For example, in a possible action unit, the athlete flips 180° backward around the Y axis, simultaneously rotates 360 ° after landing, and separates the feet from the vehicle during the flight, and the vehicle body rotates two revolutions. The standard action code obtained after evaluation of the action unit is "6434".

[0068] It should be noted that although the rotation axes of the athletes are different, the number of rotations around the rotation axes can be represented by the same number. For example, 360° rotation around the X axis corresponds to the first digit "4", and 360° rotation around the Y axis corresponds to the first digit "1" or "2", but in both cases the third digit can be represented by "2". Of course, because the difficulty of the athletes' movements around the X axis and around the Y axis is different, even if the third digit is the same, because it belongs to different motion groups, the corresponding standard motion code is also different. Therefore, the motion difficulty can be well distinguished and determined, so as to give different standard scores.

[0069] It should be noted that the third digit represents the change of direction of the athlete, and for the first group and the second group, it refers to the angle of change of the athlete's body around the Y axis from the start of the flight to the completion of the landing; for the third group, it refers to the angle of change of the athlete's body around the Z axis from the start of the flight to the completion of the landing; for the fourth group, it refers to the angle of change of the athlete's body around the X axis from the start of the flight to the completion of the landing; for the fifth group and the sixth group, it refers to the sum of the angles of change of the athlete's body around the Y axis and around the Z axis from the start of the flight to the completion of the landing; for the seventh group and the eighth group, it refers to the sum of the angles of change of the athlete's body around the Y axis and around the X axis from the start of the flight to the completion of the landing; for the ninth group, it refers to the sum of the angles of change of the athlete's body around the X axis, Y axis and Z axis from the start of the flight to the completion of the landing.

[0070] It should be noted that when the second digit is 0, it represents no separation between the person and the vehicle, and at this time the number of turns of the athlete is equal to the number of turns of the small wheel vehicle, that is, the third digit and the fourth digit are the same; when the second digit is 1 or 2, the fourth digit refers to the number of turns of the handlebar; when the second digit is 3 or 4, the fourth digit refers to the number of turns of the vehicle body; when the second digit is 5, the fourth digit refers to the sum of the number of turns of the handlebar and the vehicle body.

[0071] It should be noted that when the second digit is 0, it represents no separation between the person and the vehicle, and at this time the number of turns of the athlete is equal to the number of turns of the small wheel vehicle, that is, the third digit and the fourth digit are the same; when the second digit is 1 or 2, the fourth digit refers to the number of turns of the handlebar; when the second digit is 3 or 4, the fourth digit refers to the number of turns of the vehicle body; when the second digit is 5, the fourth digit refers to the sum of the number of turns of the handlebar and the vehicle body.

[0072] In one embodiment, a standard score corresponding to each standard motion code is set in advance, and a corresponding standard score is given to each motion unit according to the standard motion code of each motion unit. Because the difficulty of each scoring unit is different, the corresponding standard score is also different, and the higher the difficulty coefficient of the motion, the higher the preset standard score.

[0073] In one embodiment, the action parameters of each action unit are weighted and summed with the standard score to calculate the action score of each action unit, and the action scores of each action unit are added to obtain the comprehensive score of the target object. The action parameters of each action unit include the take-off height, the hang time, the action continuity parameter and the action completeness parameter, the weight of each action parameter is pre-set, the standard score of the current action unit is calculated by weighting the action parameters and the pre-set weight, and the action score of the current action unit is obtained. The action scores of all action units in the video stream of the target object competition are added to obtain the comprehensive score of the target object.

[0074] For example, one possible action unit score is 3124H, the pre-set standard score is 11 points, the action parameters are: the take-off height is 110% of the pre-set standard height, the hang time is 90% of the pre-set standard time, the action continuity parameter is 1, and the action completeness parameter is 0.76, the pre-set weight of each action parameter is: the take-off height is 20%, the hang time is 20%, the action continuity parameter is 30%, and the action completeness parameter is 30%, and the calculation method of the action score of the current action unit is (1.1×20%+0.9×20%+1×30%+0.76×30%)×11, which is 10.208.

[0075] For example, in one possible video stream of a freestyle scooter competition, the athlete performs three take-off actions in turn, the standard action codes of the action units are 3124H, 2412 and 6434, and the corresponding pre-set standard scores are 11 points, 9 points and 17 points respectively. According to the pre-set weights of the take-off height of 20%, the hang time of 20%, the action continuity parameter of 30% and the action completeness parameter of 30%, if the action scores of each action unit after weighting the action parameters are 10.208, 9.252 and 17.476 respectively, the final score of the athlete is 36.936.

[0076] Please refer to Figure 2 One embodiment of the present application provides a scoring method for a freestyle scooter competition, which is applied to the above-mentioned system for a freestyle scooter competition, and the steps of the method are as follows:

[0077] S1: acquiring the video stream of the competition of the target object through the image acquisition module to analyze one or more action units of the target object in the video stream and acquire a plurality of video frames in each action unit;

[0078] S3: determining a plurality of action parameters of the target object in each action unit based on the action units and the plurality of video frames of the action units through the image analysis module, and transmitting the action parameters to the host computer module.

[0079] S5: The host computer module inputs the plurality of action parameters into a preset scoring architecture model, the scoring architecture model performs weighted calculation on the plurality of action parameters, obtains a comprehensive score of each action unit based on the weighted calculation result, and transmits the comprehensive score to a display module;

[0080] S7: The display module receives the comprehensive score and displays the comprehensive score on the display screen.

[0081] In the embodiment, first, a video stream of a competition of a target object is acquired, taking a competition of a freestyle scooter as an example, first, a video stream of a competition of an athlete is acquired and divided into a plurality of action units, and a movement process in which the athlete performs one-time flight is an action unit. Then, each action parameter of the athlete in each action unit is acquired, the action parameter includes flight height, air time, action continuity parameter and action completeness parameter, and a standard action code of each action unit is acquired, and a standard score of each action unit is determined according to the standard action code. Finally, the standard score of each action unit is weighted according to the action parameter of each action unit, and an action score of each action unit is obtained according to the weighted operation, and the action score of each unit is added to obtain a comprehensive score of the athlete in the current competition video stream. In addition, the comprehensive score can be displayed through a display device for reference by the athlete and the referee.

[0082] The specific implementation of the method embodiment of the present application can refer to the description of the foregoing system embodiment, and the method embodiment can also achieve the technical effects of the system embodiment, which will not be described here.

[0083] The technical scheme provided by one or more embodiments of the present application acquires a video stream of a competition of a target object, automatically scores the competition process of the target object according to the video stream, and improves the scoring speed and scoring accuracy of the freestyle scooter competition. Among them, an athlete is selected as a target object, a video stream of the competition of the athlete is acquired through an image acquisition module, and the video stream is divided into different action units through an image analysis module, a host computer module analyzes and evaluates each action unit, obtains a plurality of action parameters of each action unit and a standard score of each action unit, and then calculates a comprehensive score of the athlete in the video stream, and the comprehensive score is taken as the competition result of the athlete.

[0084] It can be seen that the technical scheme provided by one or more embodiments of the present application can quickly and accurately obtain the competition results of the freestyle scooter athletes. Since the competition video of the athletes is acquired by the image acquisition module, the camera can more comprehensively detect the competition venue, and the view of the camera on the competition venue is less limited, and the competition actions of the athletes are recognized by processing and analyzing the video stream, which can avoid the situation that the score error and the too long consideration time of the score caused by the view blockage and the difficult-to-see action details of the judges, and the athletes are scored more quickly and accurately.

[0085] The embodiment of the present application also provides a scene example of the freestyle scooter competition. The scoring system, method and equipment for the freestyle scooter competition provided by the present application can be applied to the above-mentioned scene example of the freestyle scooter competition.

[0086] In actual application, the duration of one competition of the athletes is usually 60s, and the venue is provided with obstacles including walls, flying platforms and double slopes to assist the athletes to perform the air action. When the athletes start to move, the 60s countdown is started, and the athletes will be scored when completing any air action. The standard of judging the end of the action is that both wheels are in contact with the horizontal plane, and one air action is one action unit. Usually, the athletes can complete 5-6 air actions in one competition, that is, 5-6 action units.

[0087] In the embodiment, each action unit has a corresponding standard action module. In the freestyle scooter competition, common action types include “dragon tail”, “air level” and “superman”. The action meaning of “dragon tail” is that the athletes keep the body and handlebar fixed in the air, rotate the front axle of the car under the body for one circle and ride back to the seat and land, the action meaning of “air level” is that the athletes make the car body parallel to the ground in the air and then put the car body back to land, and the action meaning of “superman” is that the athletes make the body and legs as far as possible to stretch straight to the back of the car in the air. Each action type has a corresponding standard action module, and the standard action module includes a plurality of specified standard actions arranged in time.

[0088] In the embodiment, first, the image acquisition module obtains the video stream of the athlete in 60s competition time, which can be collected in real time through a live camera, or can be called through an API interface network video. Then, the image analysis module divides the above-mentioned video stream into different action units according to the action process of the athlete, and obtains a plurality of action parameters of each action unit. Through the host computer module, the target detection and target tracking of different action units are carried out, the action type of the athlete in an action unit is determined, and the action type is coded according to the preset standard, for example, "dragon tail", the standard score of the action unit is determined to be 6 points according to the standard action code, and the action score of the action unit is calculated by weighting according to the action parameters and the standard score of the action unit. According to the above-mentioned mode, the action scores of all action units in 60s competition time are calculated, and all the action scores are added to obtain the comprehensive score of the athlete, that is, the score of the athlete in this round of competition.

[0089] The above is only a scene example provided by the specification, and does not limit the present application. Any modification, equivalent replacement, etc. within the spirit and principles of the present application shall be included in the protection scope of the present application.

[0090] The scoring system of the free-style small wheeled vehicle competition in the embodiment of the present application is presented in the form of functional units. The unit here refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and a memory executing one or more software or fixed programs, or other devices that can provide the above functions.

[0091] Please refer to Figure 3 , Figure 3 is a structural schematic diagram of a computer device provided by the embodiment of the present application, as Figure 3 shown, the computer device includes one or more processors 10, a memory 20, and an interface for connecting various components, including a high-speed interface and a low-speed interface. Various components are communicatively connected to each other by different buses, and can be installed on a common motherboard or in other ways as needed. The processor can process instructions executed in the computer device, including instructions stored in the memory or memory to display GUI on an external input / output device (such as a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used with multiple memories and multiple memories. Similarly, multiple computer devices can be connected, each providing part of the necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system). Figure 3 In the embodiment, the processor 10 is taken as an example.

[0092] The processor 10 can be a central processing unit, a network processing unit, or a combination thereof. The processor 10 can further include a hardware chip. The hardware chip can be an application specific integrated circuit, a programmable logic device, or a combination thereof. The programmable logic device can be a complex programmable logic device, a field programmable logic device, a general array logic, or any combination thereof.

[0093] The memory 20 stores instructions executable by the at least one processor 10 to cause the at least one processor 10 to perform the methods illustrated in the above embodiments.

[0094] The memory 20 can include a program storage area and a data storage area. The program storage area can store an operating system and application programs required by at least one function. The data storage area can store data created according to the use of the computer device, and the like. In addition, the memory 20 can include a high-speed random access memory, and can further include a non-transitory memory, such as at least one magnetic disk storage device, a flash memory device, or other non-transitory solid-state memory device. In some alternative embodiments, the memory 20 can optionally include a memory disposed remotely with respect to the processor 10, and these remote memories can be connected to the computer device through a network. Examples of the network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0095] The memory 20 can include a volatile memory, such as a random access memory, and can also include a non-volatile memory, such as a flash memory, a hard disk, or a solid state disk. The memory 20 can further include a combination of the above-mentioned types of memories.

[0096] The computer device further includes an input device 30 and an output device 40. The processor 10, the memory 20, the input device 30, and the output device 40 can be connected through a bus or other means, and are connected through a bus in FIG. X as an example.

[0097] The input device 30 can receive input digital or character information, and generate key signal inputs related to the user settings and function controls of the computer device, such as a touch screen, a keypad, a mouse, a trackpad, a touchpad, a pointing stick, one or more mouse buttons, a trackball, a joystick, and the like. The output device 40 can include a display device, an auxiliary lighting device (e.g., an LED), a tactile feedback device (e.g., a vibration motor), and the like. The display device includes, but is not limited to, a liquid crystal display, a light emitting diode, a display, and a plasma display. In some alternative embodiments, the display device can be a touch screen.

[0098] The embodiments of the present application further provide a computer readable storage medium, and the method according to the embodiments of the present application can be implemented in hardware, firmware, or recorded in a storage medium, or be implemented as computer code originally stored in a remote storage medium or a non-transitory machine readable storage medium and downloaded to a local storage medium and stored in the local storage medium, so that the method described herein can be processed by such software on a storage medium using a general purpose computer, a special purpose processor, or programmable or special hardware. The storage medium can be a magnetic disk, an optical disk, a read-only memory, a random access memory, a flash memory, a hard disk, or a solid state disk, etc. Further, the storage medium can also include a combination of the above-mentioned memories. It can be understood that the computer, the processor, the microprocessor controller, or the programmable hardware includes a storage component that can store or receive software or computer code, when the software or computer code is accessed and executed by the computer, the processor, or the hardware, the method shown in the above embodiments is implemented.

[0099] The system, module or unit illustrated in the above embodiments can be specifically implemented by a computer chip or entity, or by a product with certain functions. A typical implementation device is a computer. Specifically, the computer can be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0100] For the convenience of description, the above apparatus is described as various units respectively in terms of functions. Of course, the functions of each unit can be implemented in the same or more software and / or hardware in the implementation of the present application.

[0101] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer usable storage media (including, but not limited to, magnetic disks, CD-ROMs, optical storage devices, etc.) containing computer usable program code.

[0102] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 one or more flow or blocks Figure 1 means for functionally implementing the steps listed in the flowchart block or blocks.

[0103] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instructions which implement the function specified in the flowchart block or blocks. Figure 1 one or more flow or blocks Figure 1 means for functionally implementing the steps listed in the flowchart block or blocks.

[0104] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 one or more flow or blocks Figure 1 means for functionally implementing the steps listed in the flowchart block or blocks.

[0105] It should be further understood that the terms "comprise", "comprising", or any other variation thereof are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements in the list, but can also include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element preceded by "comprises a" does not, without more constraints, foreclose the existence of additional identical elements in the process, method, article, or apparatus that comprises the recited element.

[0106] Each of the embodiments in the present specification is described in a progressive manner, and the same or similar parts between each embodiment can be referred to each other, and each embodiment focuses on the difference from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant part can be referred to the part of the method embodiment.

[0107] The above merely provides an example of the present application, but is not intended to limit the present application. The present application can have various modifications and changes for those skilled in the art. Any modification, equivalent replacement, improvement, etc. within the spirit and principle of the present application shall be included in the scope of claims of the present application.

[0108] Although the embodiments of the present application are described with reference to the drawings, various modifications and changes can be made by those skilled in the art without departing from the spirit and scope of the present application, and such modifications and changes shall fall within the scope defined by the appended claims.

Claims

1. A scoring system for a free-style scooter competition, characterized in that, The system comprises an image acquisition module, an image analysis module, an upper computer module, and a display module, wherein the upper computer module runs a scoring architecture model. The image acquisition module is configured to acquire a video stream of a competition of a target object, parse one or more action units of the target object in the video stream, and acquire a plurality of video frames in the action units. The image analysis module is configured to determine a plurality of action parameters in each action unit based on each action unit and the plurality of video frames in the action unit. The upper computer module is configured to acquire the one or more action units, input the action units into a preset scoring architecture model, evaluate each action unit according to a plurality of action parameters of the action unit based on the scoring architecture model, and obtain a comprehensive score of the target object based on an evaluation result. The display module is configured to acquire the comprehensive score of the target object in the upper computer module and display the comprehensive score on a terminal screen. The image acquisition module comprises an image acquisition unit and an image processing unit, wherein: The image acquisition unit is configured to acquire a video stream of a competition of a target object. The image processing unit is configured to divide one or more action units of the target object according to the video stream, and acquire a plurality of video frames in one action unit, wherein the one action unit is an action process of one jump of the target object. The image analysis module comprises a height detection unit, a time detection unit, an action continuity detection unit, and an action completeness detection unit, and the action parameters comprise a jump height, a hang time, an action continuity parameter, and an action completeness parameter, wherein: The height detection unit is configured to acquire first motion data of a current action unit, and determine a jump height of the target object according to the first motion data. The time detection unit is configured to acquire second motion data of the current action unit and time stamps corresponding to each video frame, and determine a hang time of the target object according to the second motion data and the time stamps corresponding to each video frame. The action continuity detection unit is configured to acquire a plurality of video frames of the target object in one action unit and time stamps of the video frames, and determine an action continuity parameter of the target object according to the video frames and the time stamps. The action completeness detection unit is configured to acquire a standard action module, the standard action module comprising a plurality of specified standard actions, compare each action unit with a corresponding specified standard action, and determine an action completeness parameter based on a comparison result. The upper computer module comprises a basic scoring unit and a comprehensive scoring unit, and the basic scoring unit runs a scoring architecture model, wherein: The basic scoring unit is configured to acquire each action unit of the target object, input a video stream of each action unit into the scoring architecture model, evaluate each action unit according to a preset action standard based on the scoring architecture model, obtain a standard action code of each action unit based on an evaluation result, and determine a standard score of each action unit according to the standard action code of each action unit. The comprehensive score unit is configured to obtain a standard score of each action unit, perform weighted calculation on the standard score corresponding to the current action unit according to an action parameter of the action unit, obtain an action score of each action unit based on a result of the weighted calculation, and obtain a comprehensive score of the target object based on the action scores of the action units.

2. The system of claim 1, wherein, The height detection unit is configured to obtain first motion data of the current action unit, determine a take-off height of the target object according to the first motion data, and specifically include the following steps. A unified reference baseline is determined in a video frame of the current action unit, and first motion data of the target object is detected, a relative height of the target object in each video frame is determined based on the reference baseline and the first motion data, and the first motion data includes position coordinates of the target object. Based on the relative heights of the target object in the video frames, the relative height farthest from the reference baseline is determined as the take-off height.

3. The system of claim 1, wherein, The time detection unit is configured to obtain second motion data of the current action unit and time stamps corresponding to the video frames, determine a hang time of the target object according to the second motion data and the time stamps corresponding to the video frames, and specifically include the following steps. The second motion data of the current action unit is obtained, and video frames of a take-off time and a landing time of the current action unit are determined according to the second motion data, wherein the second motion data includes vertical position coordinates, speed and acceleration of the target object. The time stamps of the video frames of the take-off time and the landing time are obtained, and the hang time of the target object is calculated according to the time stamps.

4. The system of claim 1, wherein, The action continuity detection unit is specifically configured to: Obtain a plurality of video frames of the target object, and identify action identifiers of the target object in each of the video frames; According to the time stamps of the plurality of video frames, the identified action identifiers are aggregated into an action sequence; The action sequence is input into a continuity detection model for detection, and an action continuity parameter is obtained based on a detection result, wherein the continuity model is generated by training according to an action sequence sample.

5. The system of claim 4, wherein, The continuity model is generated by training according to an action sequence sample, and specifically includes the following steps. An action sequence sample is obtained, the action sequence sample includes a set of action sequences of a plurality of action types, and the set of action sequences includes a plurality of action sequences with different continuity; The action sequence sample is input into a continuity model to train the continuity model; The trained continuity model is evaluated, and the continuity model is adjusted based on an evaluation result.

6. The system of claim 1, wherein, The action completeness detection unit is specifically configured to: Obtain one or more standard action modules, and the standard action module includes a plurality of specified standard actions arranged in time; Obtain a plurality of video frames of each action unit, identify action identifiers of the video frames, and perform feature matching between the action identifiers and the specified standard actions, and determine an action completeness parameter of each action unit based on a feature matching result.

7. The system of claim 6, wherein, The action identifiers are matched with the specified standard actions based on the feature matching result to determine the completeness parameter of each action unit. extracting a first action feature set of the action identifier and a second action feature set of the specified standard action, and performing multi-dimensional feature matching on the first action feature set and the second action feature set, wherein the first action feature set comprises one or more first action features, and the second action feature set comprises one or more second action features; obtaining an action completeness parameter of each action unit through weighted calculation according to the multi-dimensional feature matching result.

8. A method of scoring a freestyle scooter competition, characterized by, The method comprises: obtaining a video stream of a competition of a target object through an image acquisition module, and parsing one or more action units of the target object in the video stream to obtain a plurality of video frames in each action unit; determining a plurality of action parameters of the target object in each action unit through an image analysis module based on the action units and the plurality of video frames in the action units, and transmitting the action parameters to an upper computer module; the upper computer module inputs the plurality of action parameters into a preset scoring architecture model, the scoring architecture model evaluates each action unit according to the plurality of action parameters of each action unit, obtains a comprehensive score of the target object based on the evaluation result, and transmits the comprehensive score to a display module; the display module receives the comprehensive score and displays the comprehensive score to a terminal screen; The image acquisition module comprises an image acquisition unit and an image processing unit, the image acquisition unit is used to obtain a video stream of a competition of a target object, and the image processing unit is used to divide one or more action units of the target object according to the video stream and obtain a plurality of video frames in one action unit, wherein the one action unit is the action process of one jump of the target object. The image analysis module comprises a height detection unit, a time detection unit, an action continuity detection unit and an action completeness detection unit, the action parameters comprise a jump height, a hang time, an action continuity parameter and an action completeness parameter, the height detection unit is used to obtain first motion data of a current action unit and determine the jump height of the target object according to the first motion data, the time detection unit is used to obtain second motion data of the current action unit and time stamps corresponding to each video frame, and determine the hang time of the target object according to the second motion data and the time stamps corresponding to each video frame, the action continuity detection unit is used to obtain a plurality of video frames of the target object in one action unit and time stamps of the video frames, and determine the action continuity parameter of the target object according to the video frames and the time stamps, and the action completeness detection unit is used to obtain a standard action module, the standard action module comprises a plurality of specified standard actions, and each action unit is compared with the corresponding specified standard action, and the action completeness parameter is determined based on the comparison result. The upper computer module comprises a basic scoring unit and a comprehensive scoring unit. The basic scoring unit runs a scoring framework model. The basic scoring unit is configured to acquire each action unit of a target object, input a video stream of each action unit into the scoring framework model, evaluate each action unit according to a preset action standard, obtain a standard action code of each action unit based on an evaluation result, and determine a standard score of each action unit according to the standard action code of each action unit. The comprehensive scoring unit is configured to acquire the standard score of each action unit, perform weighted calculation on the standard score corresponding to a current action unit according to an action parameter of each action unit, obtain an action score of each action unit based on a weighted calculation result, and obtain a comprehensive score of the target object based on the action score of each action unit.

9. A computer device, comprising: Comprise: A memory and a processor, which are in communication connection with each other, and the memory stores computer instructions. The processor executes the computer instructions to perform the scoring method of the freestyle scooter competition according to claim 8.

Images