Ball technique and tactical ability monitoring method and device and storage medium

Abstract

The invention relates to a ball skill and tactical ability monitoring method and device and a storage medium. The method comprises the following steps: acquiring an original single-view video comprising a plurality of image frames; based on each image frame, marking a preset court key point 2D coordinate corresponding to a pre-obtained same-group court key point 3D coordinate, and optimizing and solving camera interior and exterior orientation parameters; respectively acquiring a ball 2D sequence and a human body key point 2D sequence by using different algorithms; obtaining a corresponding ball 3D position sequence and a player 3D position sequence based on the optimized camera interior and exterior orientation parameters, the ball 2D sequence and the human body key point 2D sequence; acquiring a frame rate, a ball 3D position sequence and a player 3D position sequence by using a video acquired in advance, and calculating technical and tactical key data; based on the original single-view video, the technical and tactical key data are visually presented. Compared with the prior art, the method has the advantages that the hardware environment building requirement and the information processing load are reduced, and the analysis accuracy of the player / ball 3D spatio-temporal characteristics is improved.

Description

Technical Field

[0001] The present invention relates to the field of computer vision technology, and in particular to a method, device and storage medium for monitoring ball game technical and tactical capabilities. Background Art

[0002] As a globally popular sport, ball sports place high demands on athletes’ technical and tactical abilities as well as their quick responses to game situations. In high-level competitions, subtle differences in technique and tactics often determine the outcome of a game. Therefore, accurate monitoring and analysis of players’ technical and tactical abilities are crucial to improving training efficiency and competition performance.

[0003] Traditional ball training and game analysis methods rely on the observation of coaches and analysts. This method is highly subjective, inefficient, and difficult to achieve large-scale and in-depth data analysis. With the development of computer vision technology, computers can accurately identify and track objects by analyzing images and videos, and have been widely used in various fields. In sports analysis, computer vision technology can provide an efficient and objective means to support the monitoring and analysis of athlete performance. Specifically, the key technologies of ball positioning and tracking and human posture detection and tracking based on computer vision make it possible to monitor and analyze technical and tactical capabilities. For example, Chinese patent CN116958872A discloses an intelligent auxiliary training method and system for badminton. The method first uses the court feature points in the artificially marked RGB image to correspond to the real court position in the world coordinate system, thereby calculating the internal and external orientation parameters of the camera. For the badminton 2D positioning task, the invention uses the current mainstream deep learning algorithm to estimate the heat map of the center position of the badminton on the 2D image, and obtains the 2D pixel coordinates of the badminton by weighted summation. For the task of estimating the player's 2D posture, the invention adopts a top-down design and completes it using the target detection-target tracking-posture recognition method. Specifically, the target detection algorithm is first used to obtain the bounding boxes of all identifiable human targets, and the court sidelines are used for filtering to retain the corresponding bounding boxes of the players. Then the 2D tracking algorithm is used to obtain the corresponding bounding box sequence for each player. The image in the bounding box is cropped and enlarged, and the 2D posture estimation algorithm is used to obtain the player's 2D posture. After that, triangulation is performed based on the 2D information of the badminton and the player in the monocular or binocular image coordinate system and the camera parameters to obtain the 3D position of the badminton and the player, and continuous images or videos are used to form the movement trajectory of the badminton and the player. However, the existing methods still have the problems of high requirements for hardware environment construction and high information processing load. Summary of the invention

[0004] The purpose of the present invention is to provide a method, device and storage medium for monitoring ball game technical and tactical capabilities in order to overcome the defects of the above-mentioned prior art in terms of high requirements for hardware environment construction and high information processing load.

[0005] The purpose of the present invention can be achieved by the following technical solutions:

[0006] According to a first aspect of the present invention, a method for monitoring ball game technical and tactical capabilities is provided, comprising the following steps: obtaining an original single-view video including a plurality of image frames; based on each image frame, optimizing and solving the internal and external position parameters of the camera by corresponding the 2D coordinates of key points of a court that are pre-set by marking with the 3D coordinates of key points of the court that are pre-acquired in the same group; based on each image frame, obtaining a 2D sequence of balls and a 2D sequence of key points of a human body by using different algorithms; based on the optimized internal and external position parameters of the camera, the 2D sequence of balls and the 2D sequence of key points of a human body, obtaining the corresponding 3D position sequence of balls and the 3D position sequence of players; obtaining the frame rate of the video obtained in advance, the 3D position sequence of balls and the 3D position sequence of players, and calculating the key technical and tactical data; based on the original single-view video, visually presenting the key technical and tactical data.

[0007] As a preferred technical solution, the process of optimizing the camera's internal and external orientation parameters specifically includes: selecting a first image frame, initializing the 2D coordinates of the first key point that marks the first image frame, and using the first key point 2D coordinates and the corresponding 3D coordinates of the court key point to calculate the corresponding camera's internal and external orientation parameters; based on the first key point 2D coordinates, when the camera moves, using a deep learning-based tracking and matching algorithm to obtain the second key point 2D coordinates of the remaining image frames, and using the second key point 2D coordinates of each image frame and the corresponding 3D coordinates of the court key point to calculate the corresponding camera's internal and external orientation parameters.

[0008] As a preferred technical solution, the tracking and matching algorithm includes the LightGlue algorithm.

[0009] As a preferred technical solution, different algorithms are used to obtain the 2D sequence of balls and the 2D sequence of human key points respectively. The specific process includes: using the TrackNetv3 algorithm to obtain the 2D sequence of balls; using the RTMPose algorithm to obtain the 2D key point coordinate sequences of multiple players, and using the OC-SORT algorithm to perform multi-target tracking on the 2D key coordinate sequence of each player to obtain the final 2D sequence of human key points.

[0010] As a preferred technical solution, the corresponding 3D position sequence of balls and the 3D position sequence of players are obtained. The specific process includes: selecting a key ball and marking the start image frame, end image frame, ball depth and player depth of the key ball; calculating the corresponding 3D position sequence of balls and the 3D position sequence of players based on the ball depth, the player depth, the optimized internal and external orientation parameters of the camera, the ball 2D sequence and the human body key point 2D sequence; and obtaining the 3D trajectory of the key ball by using kinematic theory modeling.

[0011] As a preferred technical solution, the tactical key data include net crossing speed, attack angle and defense distance.

[0012] As a preferred technical solution, based on the 3D trajectory of the key ball, the ball landing point is estimated; the time and corresponding position of the ball trajectory on both sides of the net are obtained; based on the time, the corresponding position and the frame rate of the video, the net crossing speed is calculated; the offensive player position and the defensive player position in the starting image frame are obtained, and then a first connecting line between the offensive player position and the defensive player position, and a second connecting line between the offensive player position and the ball landing point are obtained; based on the first connecting line and the second connecting line, the offensive angle is calculated; based on the defensive player position and the ball landing point, the defensive distance is calculated.

[0013] As a preferred technical solution, visually presenting the key technical and tactical data specifically includes: using Unity to add special effects to the key technical and tactical data, and using an augmented reality method to add it to the original single-view video, and displaying it at a corresponding position through a display module.

[0014] According to a second aspect of the present invention, there is provided a device for monitoring ball technical and tactical capabilities, the device being used to implement the method, the device comprising a court 3D calibration module, a ball and player positioning module, a technical and tactical analysis module and a 3D rendering module, wherein: the court 3D calibration module is used to obtain an original single-view video including a plurality of image frames, and based on each image frame, uses a marker to correspond a pre-set 2D coordinate of a court key point with a pre-acquired 3D coordinate of a same group of court key points, and optimizes and solves the camera's internal and external orientation parameters; the ball and player positioning module is used to obtain a ball 2D sequence and a human body key point 2D sequence respectively based on each image frame using different algorithms, and obtains a corresponding ball 3D position sequence and a player 3D position sequence based on the optimized camera's internal and external orientation parameters, the ball 2D sequence and the human body key point 2D sequence; the technical and tactical analysis module is used to obtain a frame rate, the ball 3D position sequence and the player 3D position sequence using the pre-acquired video, and calculate technical and tactical key data; the 3D rendering module is used to visually present the technical and tactical key data based on the original single-view video.

[0015] According to a third aspect of the present invention, there is provided a storage medium having a program stored thereon, wherein the program implements the method described above when executed.

[0016] Compared with the prior art, the present invention has the following beneficial effects:

[0017] 1. The ball game technical and tactical ability monitoring method provided by the present invention only requires a single-view video as input, and does not require hardware such as wearable devices and special cameras. It uses key points of the court combined with multiple algorithms to achieve high-precision 3D calibration of the court, obtain ball and player spatial information, and visualize the key data of technical and tactical ability. It can effectively reduce the requirements for hardware environment construction and information processing load, and at the same time provide accurate and intuitive players' technical and tactical abilities, providing convenience for monitoring and analyzing players' technical and tactical abilities.

[0018] 2. The present invention provides a 3D stadium calibration process based on a single perspective, using the stadium markings as a reference, and optimizing the internal and external orientation parameters of the camera through interactively marking the 2D coordinates of key points and a tracking and matching algorithm based on deep learning, which can support the calibration of dynamic cameras and effectively improve the accuracy of stadium calibration;

[0019] 3. The present invention provides a 3D positioning process for players and balls based on a single perspective, and uses ball depth, player depth, optimized camera internal and external orientation parameters, ball 2D sequence and human key point 2D sequence to calculate the corresponding ball 3D position sequence and player 3D position sequence, that is, using pixel-precision position and joint-point granularity player posture information, supplemented by an interactive depth annotation tool, to obtain ball and player 3D position sequences from a single perspective, and obtain multi-dimensional technical and tactical ability indicators such as net-crossing speed, attack angle and defense distance based on kinematic modeling, which is an extension of basic ball and player motion trajectory and other kinematic information, and effectively improves the accuracy and comprehensiveness of 3D spatiotemporal feature analysis of players and balls;

[0020] 4. On the basis of obtaining accurate data, the present invention uses Unity to render realistic 3D animations of ball games, supplemented by key event special effects, to provide users with an improved technical and tactical interface for analysis, and can achieve accurate and multi-dimensional playback of ball game technical and tactical indicators. Its visualization capability significantly improves the intuitiveness and convenience of technical and tactical capability monitoring and analysis;

[0021] 5. The present invention can still achieve the beneficial effects of the provided technical solution under different ball trajectory modeling methods, different human skeleton key points or different 3D rendering special effects methods, and can effectively improve the scene adaptability of the ball technical and tactical ability monitoring method and device, which is convenient for promotion. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 A schematic flow chart of the method provided by the present invention;

[0023] Figure 2 A schematic diagram of the execution flow of the device is provided for the present invention. DETAILED DESCRIPTION

[0024] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. This embodiment is implemented based on the technical solution of the present invention, and provides a detailed implementation method and specific operation process, but the protection scope of the present invention is not limited to the following embodiments.

[0025] Example 1

[0026] like Figure 1 As shown, this embodiment provides a method for monitoring ball skills and tactics, which can achieve high-precision 3D calibration of the court, accurate acquisition of ball trajectories and player postures, calculation and visualization of multi-dimensional key technical and tactical data of players (i.e., multi-dimensional technical and tactical ability indicators), and 3D animation playback without the need for hardware such as wearable devices and special cameras. The method includes the following steps:

[0027] Step S1, obtaining an original single-view video including a plurality of image frames;

[0028] Step S2, based on each image frame, using the correspondence between the 2D coordinates of the key points of the court set in advance and the 3D coordinates of the key points of the court in advance obtained in the same group, optimizing and solving the internal and external orientation parameters of the camera;

[0029] Step S3, based on each image frame, using different algorithms to respectively obtain a 2D sequence of balls and a 2D sequence of key points of a human body;

[0030] Step S4, based on the optimized camera internal and external orientation parameters, the ball 2D sequence and the human body key point 2D sequence, obtaining the corresponding ball 3D position sequence and the player 3D position sequence;

[0031] Step S5, using the pre-acquired video to obtain the frame rate, the ball 3D position sequence and the player 3D position sequence, and calculating the key technical and tactical data;

[0032] Step S6, based on the original single-view video, visually presenting key technical and tactical data.

[0033] This embodiment also provides a ball game technical and tactical ability monitoring device for implementing each step of the above method. Figure 2 The execution process is shown. The device includes a stadium 3D calibration module, a ball and player positioning module, a technical and tactical analysis module, and a 3D rendering module. Among them, the stadium 3D calibration module is used to implement steps S1-S2, the ball and player positioning module is used to implement steps S3-S4, the technical and tactical analysis module is used to implement step S5, and the 3D rendering module is used to implement step S6. The specific process of the collaborative work of each module is as follows:

[0034] (1) Stadium 3D Calibration Module

[0035] The stadium 3D calibration module interactively marks the key points of the stadium, uses the tracking and matching algorithm based on deep learning to obtain the key point positions in each image frame in the original video, solves the internal and external orientation parameters of the camera, and realizes the stadium calibration. Specifically:

[0036] This module uses Python OpenCV software library to implement an interactive graphical interface. Users mark the four corner points of the court sideline and the two upper corners of the net (if any) in turn, and input the 2D coordinates of the key points in the image coordinate system and the corresponding 3D coordinates of the key points of the court into the court 3D calibration module. By matching the 3D coordinates of the key points of the court known in the world coordinate system with the 2D coordinates of the key points in the image coordinate system marked by the user, the internal and external orientation parameters of the camera can be optimized.

[0037] For the case of camera movement, the module first initializes the 2D coordinates of the key points of the first image frame (for example, the 0th frame is selected as the first image frame), obtains the 2D coordinates of the first key points, and then corresponds them to the 3D coordinates of the corresponding key points of the court, and solves the corresponding camera internal and external orientation parameters; then, the LightGlue algorithm is used to track and match the 2D positions of the key points in the remaining image frames of the original video (that is, multiple second key point 2D coordinates are obtained), and the camera internal and external orientation parameters in the remaining image frames of the original video are calculated in the same way.

[0038] It should be noted that the key points in this embodiment refer to a series of points with specific physical meanings, such as the corner points of the court. Their 3D coordinates refer to their positions in the world coordinate system, and their 2D coordinates refer to their positions in the image coordinate system. Due to camera movement, the 2D coordinates of the key points of the court in different frames will change, so it is necessary to track and match the 2D coordinates of the key points in the remaining frames.

[0039] In practical applications, for the image frame I i Given a video stream V = {I0, I1, ...}, the stadium 3D calibration module uses the 0th frame to calibrate the camera's internal and external parameters K and E, that is, manually check the 2D coordinates of the key points of the stadium corner 2d , and the known 3D coordinates of the key point of the court corner 3d As the input of the stadium 3D calibration module, the camera's internal and external orientation parameters are optimized. If the camera moves, the LightGlue algorithm uses the 2D coordinates of the key points in the previous period Calculate the 2D coordinates of the key points at the current moment Thus, the 2D coordinates of the key points of the court in each frame are obtained.

[0040] (2) Ball and player positioning module

[0041] The ball and player positioning modules use ball and player detection and tracking algorithms based on deep learning to obtain the 2D positions of balls and players in each image frame of the original video. After that, an interactive graphical interface is used to determine the start image frame and end image frame of the key ball, annotate the ball depth and player depth, and obtain the 3D positions of the ball and players in the key frame; kinematic theory modeling is used to obtain the 3D trajectory of the key ball. Specifically:

[0042] For ball 2D positioning: using image frames as input, the TrackNetv3 algorithm fine-tuned for different motion types is used to obtain the ball 2D sequence;

[0043] For player 2D positioning: using image frames as input, the RTMPose algorithm is used to estimate 2D human posture, that is, to obtain multiple player 2D key point coordinate sequences, and then the OC-SORT algorithm is used for multi-target tracking to obtain a 2D sequence of human key points.

[0044] This module uses Python OpenCV software library to implement another interactive graphical interface for selecting key balls from image frames. The user marks the start and end image frames of the key balls, the ball depth and the player depth, and then uses the camera internal and external orientation parameters obtained in the court 3D calibration module to calculate the 3D position of the ball in the start and end image frames, and the 3D position of the player in each frame. When the 2D coordinates, depth and camera internal and external orientation parameters of the ball / player are determined, the 3D position can be calculated based on the 2D coordinates, depth and camera parameters in multiple perspectives. Considering the gravity, air resistance and Magnus force generated by the rotation of the ball during its flight in the air, the ball's motion trajectory between the start and end image frames can be accurately modeled, and the ball's landing point can be estimated.

[0045] In practical applications, the TrackNetv3 algorithm uses image frame I i As input, output the 2D coordinates of the ball The 2D coordinates of the ball in each frame form a 2D sequence of balls x ball ; RTMPose algorithm uses image frame I i As input, output the coordinates of multiple 2D key points of players in the scene Then the OC-SORT algorithm is used to calculate the key point coordinate sequence x of each player. player The optimization is performed to obtain the final 2D sequence of key points of the human body. The various algorithms used in this module have the characteristics of high accuracy and low latency, and can be used for efficient monitoring and analysis.

[0046] (3) Technical and tactical analysis module

[0047] The technical and tactical analysis module calculates key technical and tactical parameters based on the ball 3D position sequence and the player 3D position sequence. Both the ball 3D position sequence and the player 3D position sequence are time-series 3D data, including the 3D positions of balls and players at different times.

[0048] This module mainly calculates the key technical and tactical data based on the frame rate of the video and the 3D position data obtained by the ball and player positioning module. The key technical and tactical data include net speed, attack angle, and defense distance. The calculation process of each key technical and tactical data includes:

[0049] First, the ball landing point is estimated based on the 3D trajectory of the key ball obtained by the ball and player positioning module;

[0050] Secondly, the time and corresponding position of the ball trajectory on both sides of the net are obtained; and the speed of passing the net is calculated based on the time and corresponding position and the frame rate of the video. The speed of passing the net is the average speed between two frames, that is, the distance of the ball multiplied by the frame rate;

[0051] Next, the offensive player position and the defensive player position in the start image frame are obtained, and then a first connecting line between the offensive player position and the defensive player position and a second connecting line between the offensive player position and the ball landing point are obtained, and the offensive angle is calculated from the angle between the two connecting lines;

[0052] Finally, the defense distance is calculated based on the distance between the defense player's position in the starting image frame and the ball landing point.

[0053] The following are some examples:

[0054] The ball trajectory function f is a function of the initial velocity v0, initial position x0 and time t, f(v0,x0,t). Given the initial position x0, the middle point position x1 and the corresponding time t1, the initial velocity v0 is optimized and solved:

[0055] v0= argmin(||x1 – f(v0,x0,t0)||) (1)

[0056] Among them, f is obtained by integrating at each moment:

[0057] f =∫(v t +a t t)*tdt+x0 (2)

[0058] Among them, the acceleration a t It is expressed as:

[0059] a t = -g-λ d *norm(v)*v / m+λ l *norm(v)*cross(ω,v) / m (3)

[0060] Where g is the acceleration due to gravity, λ is d is the resistance empirical parameter 0.00128625, λ l is the empirical parameter of the Magnus force 0.0005145, ω is the horizontal vector, and m is the mass of the ball.

[0061] Therefore, the network speed v net It is expressed as:

[0062] v net = ||x left -x right ||2*fps (4)

[0063] In the formula, x left 、x right is the position of the balls on both sides of the net at the same time, and fps is the frame rate;

[0064] The attack angle angle is expressed as:

[0065]

[0066] In the formula, x off is the attacking player position, x def For defensive player position, The attack angle is the angle between the vector connecting the attacking player and the defensive player and the vector connecting the defensive player and the ball landing point.

[0067] The defense distance d is expressed as:

[0068]

[0069] In the formula, x def For defensive player position, The location where the ball lands.

[0070] (4) 3D rendering module

[0071] The 3D rendering module uses Unity to design visualization effects of key technical and tactical data, which are superimposed on the original video in an augmented reality way. Specifically:

[0072] This module uses Unity to add special effects to the key technical and tactical data calculated in the technical and tactical analysis module, and adds it to the original video in an augmented reality way. This module uses the start, middle, end, and expected landing point of the ball's 3D trajectory to fit the Bezier curve to control the trajectory of the arrow special effect in the virtual 3D coordinate system. The player's 3D position at each moment and the expected landing point of the key ball will be marked by a ring special effect at the corresponding calculated position. Key technical and tactical data such as net speed, attack angle, and defense distance are presented by the Unity UI construction panel, and the specific values ​​are additionally displayed in the corresponding screen position using TextMesh Pro.

[0073] In summary, the method and device for monitoring ball game skills and tactics provided in this embodiment have the following advantages:

[0074] 1) Use the key points of the stadium to calibrate the camera and support dynamic camera calibration, which has better scene generalization capabilities;

[0075] 2) Using ball dynamics modeling, combined with a deep learning-based tracking and matching algorithm and an interactive depth estimation algorithm, the 3D trajectory of the ball is obtained, which has higher accuracy for ball and player positioning;

[0076] 3) Use Unity to add special effects rendering to key technical and tactical data, with complete functions and richer technical and tactical capability presentation forms.

[0077] Example 2

[0078] The present embodiment provides a storage medium on which a program is stored, and the aforementioned method is implemented when the program is executed. The program code for implementing the method of the present invention can be written in any combination of one or more programming languages. These program codes can be provided to a processor or controller of a general-purpose computer, a special-purpose computer or other programmable data processing device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code can be executed entirely on the machine, partially on the machine, partially on the machine as an independent software package and partially on a remote machine or completely on a remote machine or server. In the context of the present invention, a computer-readable medium can be a tangible medium that can contain or store a program for use by an instruction execution system, device or equipment or used in combination with an instruction execution system, device or equipment. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the above. More specific examples of machine-readable storage media would include electrical connections based on one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), optical fibers, a portable compact disk-read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0079] The preferred specific embodiments of the present invention are described in detail above. It should be understood that a person skilled in the art can make many modifications and changes based on the concept of the present invention without creative work. Therefore, any technical solution that can be obtained by a person skilled in the art through logical analysis, reasoning or limited experiments based on the concept of the present invention on the basis of the prior art should be within the scope of protection determined by the claims.

Claims

1. A method for monitoring ball game technical and tactical abilities, characterized in that: The following steps are involved: Acquire an original single-view video including a plurality of image frames; Based on each image frame, the camera's internal and external orientation parameters are optimized by matching the 2D coordinates of the key points of the court set in advance with the 3D coordinates of the key points of the court in advance. Based on each image frame, different algorithms are used to obtain the 2D sequence of balls and the 2D sequence of key points of the human body respectively; Based on the optimized internal and external orientation parameters of the camera, the 2D sequence of the ball and the 2D sequence of key points of the human body, a corresponding 3D position sequence of the ball and a 3D position sequence of the player are obtained; Utilizing the pre-acquired video acquisition frame rate, the ball 3D position sequence and the player 3D position sequence, to calculate key technical and tactical data; Based on the original single-view video, the key technical and tactical data are visualized.

2. The method for monitoring ball game skills and tactics according to claim 1, characterized in that: The process of optimizing the camera's internal and external orientation parameters specifically includes: Select a first image frame, initialize and mark the 2D coordinates of a first key point of the first image frame, and calculate corresponding camera internal and external orientation parameters using the 2D coordinates of the first key point and the 3D coordinates of the corresponding court key point; Based on the 2D coordinates of the first key points, when the camera moves, the 2D coordinates of the second key points of the remaining image frames are obtained using a deep learning-based tracking and matching algorithm, and the corresponding internal and external orientation parameters of the camera are calculated using the 2D coordinates of the second key points of each image frame and the corresponding 3D coordinates of the key points of the court.

3. The method for monitoring ball game skills and tactics according to claim 2, characterized in that: The tracking and matching algorithm includes the LightGlue algorithm.

4. The method for monitoring ball game skills and tactics according to claim 1, characterized in that: Different algorithms are used to obtain the 2D sequence of balls and the 2D sequence of key points of the human body. The specific process includes: Use TrackNetv3 algorithm to obtain 2D sequence of balls; The RTMPose algorithm is used to obtain the 2D key point coordinate sequences of multiple players, and the OC-SORT algorithm is used to perform multi-target tracking on the 2D key coordinate sequence of each player to obtain the final 2D sequence of human body key points.

5. The method for monitoring ball game skills and tactics according to claim 1, characterized in that: Obtain the corresponding ball 3D position sequence and player 3D position sequence. The specific process includes: Select a key ball, and mark the start image frame, end image frame, ball depth, and player depth of the key ball; Calculate corresponding ball 3D position sequences and player 3D position sequences based on the ball depth, the player depth, the optimized camera internal and external orientation parameters, the ball 2D sequence and the human body key point 2D sequence; Use kinematics theory modeling to obtain the 3D trajectory of the key ball.

6. The method for monitoring ball game skills and tactics according to claim 5, characterized in that: The key tactical data include net speed, attack angle and defense distance.

7. The method for monitoring ball game skills and tactics according to claim 6, characterized in that: The calculation process of calculating the key technical and tactical data includes: Based on the key ball 3D trajectory, estimate the ball landing point; Obtain the time and corresponding position of the ball trajectory on both sides of the net; Calculating the network speed based on the time, the corresponding position and the frame rate of the video acquisition; Acquire the offensive player position and the defensive player position in the start image frame, and then obtain a first connecting line between the offensive player position and the defensive player position, and a second connecting line between the offensive player position and the ball landing point; Calculating the attack angle based on the first connecting line and the second connecting line; The defense distance is calculated based on the defense player's position and the ball landing point.

8. The method for monitoring ball game skills and tactics according to claim 1, characterized in that: Visualize the key technical and tactical data, including: Special effects are added to the key technical and tactical data using Unity, and are added to the original single-view video using an augmented reality method, and are displayed at a corresponding position through a display module.

9. A ball game technical and tactical ability monitoring device, characterized in that: The device is used to implement the method according to any one of claims 1 to 8, and the device includes a stadium 3D calibration module, a ball and player positioning module, a technical and tactical analysis module, and a 3D rendering module, wherein: The court 3D calibration module is used to obtain an original single-view video including a plurality of image frames, and based on each image frame, optimize and solve the camera internal and external orientation parameters by marking the 2D coordinates of the court key points pre-set by the markers and the 3D coordinates of the same group of court key points pre-acquired; The ball and player positioning module is used to obtain a ball 2D sequence and a human key point 2D sequence based on each image frame using different algorithms, and obtain a corresponding ball 3D position sequence and a player 3D position sequence based on the optimized camera internal and external orientation parameters, the ball 2D sequence and the human key point 2D sequence; The technical and tactical analysis module is used to calculate key technical and tactical data by using the pre-acquired video acquisition frame rate, the ball 3D position sequence and the player 3D position sequence; The 3D rendering module is used to visualize the key technical and tactical data based on the original single-view video.

10. A storage medium having a program stored thereon, characterized in that: When the program is executed, the method according to any one of claims 1 to 8 is implemented.

Images