Device and method for analyzing boxing actions to provide boxing training suggestions

Abstract

A device and a method for analyzing boxing actions to provide boxing training suggestions are provided, which analyze historical competition images of competition opponents to generate opponent action information of the competition opponents, and simulate action strategies and action modes of the competition opponents by using a machine learning model trained according to the opponent action information of the competition opponents. According to the action strategy and the action mode, through a reality technology, a user and a virtual opponent are provided to carry out a fighting exercise, a fighting action of the user in the fighting exercise is analyzed to generate target action information, and a training suggestion corresponding to a competition opponent is generated according to the target action information. Therefore, the reaction can be made in real time according to the action of the user, improvement suggestions can be provided according to the expression of the user, and the technical effects of truly simulating the reaction of the opponent and providing integral improvement suggestions are achieved.

Description

Technical Field

[0001] A training suggestion providing device and method thereof, in particular, a device and method for analyzing boxing movements to provide boxing training suggestions. Background Art

[0002] Boxing is a sport in which two players fight each other. In the sport, the two players can only use their fists to attack and defend. Usually, the players need to wear boxing gloves to reduce the extent of injuries to the opponent.

[0003] Previously, the training of boxing players mainly relied on the experience of coaches. In most cases, the training results varied significantly with the coach's experience in competition and teaching. Therefore, with the increase in processing efficiency and the advancement of technology, the training methods of boxing players have also begun to turn to scientific training, such as using training systems that can simulate fighting with opponents. The training system can usually capture the user's movements and perform posture corrections.

[0004] However, in most current training systems, the simulated virtual opponents are mainly used to practice fighting with the user based on a pre-set fixed action sequence. Therefore, the virtual opponent cannot react to the user's behavior in real time, lacks flexibility, realism and presence, and limits the user's ability to cope with real games. At the same time, although most training systems also provide analysis functions, most training systems only provide correction feedback for each single action of the user, and cannot provide suggestions for the overall game and the actual opponent's actions. The user cannot effectively obtain improvement suggestions for coping with the opponent.

[0005] In summary, it can be seen that the existing technology has long had the problem that the boxing training system cannot respond immediately to the user's actions and cannot provide improvement suggestions to the user based on the user's performance. Therefore, it is necessary to propose improved technical means to solve this problem. Summary of the invention

[0006] In view of the problem that the boxing training system in the prior art cannot respond to the user's movements in real time and cannot provide improvement suggestions to the user based on the user's performance, the present invention discloses a device and method for analyzing boxing movements to provide boxing training suggestions, wherein:

[0007] The device disclosed in the present invention for analyzing boxing movements to provide boxing training suggestions at least comprises: a data acquisition module for acquiring historical game images, wherein the historical game images include game opponents; a data analysis module for analyzing the historical game images to generate opponent action information of the game opponents, wherein the opponent action information includes punching or defensive movements, positions, steps and moving speeds; an opponent simulation module for training a machine learning model based on the historical game images and the opponent action information, and for using the machine learning model to generate action strategies and action patterns of virtual opponents that simulate the game opponents, wherein the action strategies include attack or defense, steps and moving speeds, and the action patterns include punching frequency, methods, speed, strength or blocking positions; a reality interaction module for providing users with fighting movements and virtual opponents for fighting practice through reality technology based on the action strategies and action patterns, so that the data acquisition module acquires fighting movement data corresponding to the fighting movements, and the data analysis module is further used to analyze the fighting movement data to obtain target movement information of the user, wherein the target movement information includes attack information, defense information and moving information; and a suggestion generation module for generating training suggestions corresponding to the game opponents based on the target movement information.

[0008] The method disclosed in the present invention for analyzing boxing movements to provide boxing training suggestions comprises at least the following steps: obtaining historical game images, the historical game images including game opponents; analyzing the historical game images to generate opponent action information of the game opponents, the opponent action information including punching or defensive movements, positions, footwork and moving speed; training a machine learning model based on the historical game images and the opponent action information; using the machine learning model to generate action strategies and action patterns of a virtual opponent simulating the game opponents, the action strategies including attack or defense, footwork, moving speed, and the action patterns including punching frequency, method, speed, strength or blocking position; providing a user with fighting movements and practicing fighting with a virtual opponent through reality technology based on the action strategies and action patterns; obtaining fighting movement data corresponding to the fighting movements; analyzing the fighting movement data to obtain the user's target action information, the target action information including attack information, defense information, and moving information; and generating training suggestions corresponding to the game opponent based on the target action information.

[0009] The device and method disclosed in the present invention are as described above. The difference between the present invention and the prior art is that the present invention generates opponent action information of the opponent by analyzing the historical game images of the opponent, and uses the machine learning model trained based on the opponent action information of the opponent to simulate the action strategy and action pattern of the opponent, and then provides the user with a virtual opponent to practice fighting with the virtual opponent through reality technology based on the action strategy and action pattern, analyzes the user's fighting actions in the fighting practice to generate target action information, and generates training suggestions corresponding to the opponent based on the target action information, thereby solving the problems existing in the prior art and achieving the technical effect of truly simulating the opponent's reaction and providing overall improvement suggestions. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 Schematic diagram of components of a device for analyzing boxing movements to provide boxing training suggestions according to the present invention.

[0011] Figure 2 A schematic diagram of a module of the processor provided in the present invention.

[0012] Figure 3 The present invention is a flowchart of the method for analyzing boxing movements to provide boxing training suggestions.

[0013] Description of reference numerals:

[0014] 100: Installation

[0015] 110: Memory

[0016] 120: Camera

[0017] 130: Communication interface

[0018] 140: Storage Media

[0019] 150: Input and output components

[0020] 170: Processor

[0021] 190: Bus

[0022] 210: Data acquisition module

[0023] 230: Data Analysis Module

[0024] 250: Opponent Simulation Module

[0025] 270: Reality Interaction Module

[0026] 290: Suggestion generation module

[0027] 400: Reality Device

[0028] Step 310: Obtain historical game images including game opponents

[0029] Step 320: Analyze historical game images to generate opponent action information of game opponents

[0030] Step 330: Train the machine learning model based on the opponent's action information

[0031] Step 340: Generate action strategies and action patterns of virtual opponents simulating game opponents using machine learning models

[0032] Step 350: Provide the user with fighting movements and practice fighting with a virtual opponent through real-life technology based on the action strategy and action mode

[0033] Step 360: Obtain fighting action data corresponding to the fighting action

[0034] Step 370: Analyze the fighting action data to obtain the user's target action information

[0035] Step 380: Generate training suggestions corresponding to the opponent based on the target action information DETAILED DESCRIPTION

[0036] The features and implementation methods of the present invention will be described in detail below with reference to the accompanying drawings and embodiments. The content is sufficient to enable any person skilled in the art to easily and fully understand the technical means used by the present invention to solve the technical problems and implement them accordingly, thereby achieving the effects that can be achieved by the present invention.

[0037] The present invention can simulate a boxer to generate a virtual opponent, and generate training suggestions based on the fighting practice between the virtual opponent and the user through the reality technology. Among them, the reality technology mentioned in the present invention includes augmented reality (AR), virtual reality (VR), and mixed reality (MR).

[0038] The apparatus for implementing the present invention may be a computing device. The computing device mentioned in the present invention includes but is not limited to hardware components such as one or more processing modules, one or more memory modules, and a bus connecting different hardware components (including memory modules and processing modules). Through the multiple hardware components included, the computing device can load and execute an operating system so that the operating system runs on the computing device, and can also execute software or programs. The computing device also includes a housing, and the above-mentioned hardware components are arranged in the housing.

[0039] The bus of the computing device mentioned in the present invention may include one or more types, such as a data bus, an address bus, a control bus, an expansion bus, and / or a local bus. The bus of the computing device includes, but is not limited to, an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, a Video Electronics Standards Association (VESA) local bus, a serial Universal Serial Bus (USB), a Peripheral Component Interconnect Express (PCI Express, PCI-E / PCIe) bus, and the like.

[0040] The processing module of the computing device of the present invention is coupled to a bus. The processing module includes a register group or a register space, and the register group or the register space can be completely set on the processing chip of the processing module, or all or part of it can be set outside the processing chip and coupled to the processing chip via a dedicated electrical connection and / or via a bus. The processing module can be a central processing unit, a microprocessor or any suitable processing element. If the computing device is a multi-processor device, that is, the computing device includes multiple processing modules, the processing modules included in the computing device are the same or similar, and are coupled and communicated through a bus. In some embodiments, the processing module can interpret a computer instruction or a series of multiple computer instructions to perform specific calculations or operations, such as mathematical operations, logical operations, data comparison, copying / moving data, etc., so as to drive other hardware components in the computing device or run an operating system or execute various programs and / or modules. The computer instructions may be assembly language instructions, instruction set architecture instructions, machine instructions, machine-dependent instructions, microinstructions, firmware instructions, or source code or object code written in any combination of one or more programming languages, and the computer instructions may be executed completely on a single computing device, partially on a single computing device, partially on one computing device and partially on another connected computing device. The above-mentioned programming languages ​​include object-oriented programming languages, such as Common Lisp, Python, C++, Objective-C, Smalltalk, Delphi, Java, Swift, C#, Perl, Ruby, etc., and conventional procedural programming languages, such as C language or other similar programming languages.

[0041] The computing device also usually includes one or more chipsets. The processing module of the computing device can be coupled to the chipset or electrically connected to the chipset through a bus. The chipset is composed of one or more integrated circuits (ICs), including a memory controller and a peripheral input / output (I / O) controller, etc. That is, the memory controller and the peripheral input / output (I / O) controller can be included in one integrated circuit, or can be implemented using two or more integrated circuits. The chipset usually provides input / output and memory management functions, as well as multiple general and / or special registers, timers, etc., wherein the above-mentioned general and / or special registers and timers can be accessed or used by one or more processing modules coupled or electrically connected to the chipset. In some embodiments, the chipset may also be part of the processing module.

[0042] The processing module of the computing device can also access the data in the memory module and the large-capacity storage area installed on the computing device through the memory controller. The above-mentioned memory module includes any type of volatile memory and / or non-volatile memory (NVRAM) memory, such as static random access memory (SRAM), dynamic random access memory (DRAM), read-only memory (ROM), flash memory, etc. The above-mentioned large-capacity storage area can include any type of storage device or storage medium, such as a hard disk drive, an optical disc, a flash drive, a memory card, a solid-state disk (SSD), or any other storage device. In other words, the memory controller can access the data in the static random access memory, the dynamic random access memory, the flash memory, the hard disk drive, and the solid-state disk.

[0043] The processing module of the computing device can also be connected and communicated with peripheral devices or interfaces such as peripheral output devices, peripheral input devices, communication interfaces, various data or signal receiving devices, etc. through the peripheral input / output controller via the peripheral input / output bus. The peripheral input device can be any type of input device, such as a keyboard, a mouse, a trackball, a touchpad, a joystick, etc. The peripheral output device can be any type of output device, such as a display, a printer, etc. The peripheral input device and the peripheral output device can also be the same device, such as a touch screen, etc. The communication interface can include a wireless communication interface and / or a wired communication interface. The wireless communication interface can include an interface supporting a wireless local area network (such as Wi-Fi, Zigbee, etc.), Bluetooth, infrared, near-field communication (NFC), a mobile communication network (cellular network) such as 3G / 4G / 5G, or other wireless data transmission protocols. The wired communication interface can be an Ethernet device, a DSL modem, a cable modem, an asynchronous transfer mode (ATM) device, or an optical fiber communication interface and / or component, etc. The data or signal receiving device may include a GPS receiver or a physiological signal receiver, and the physiological signals received by the physiological signal receiver include but are not limited to heartbeat, blood oxygen, etc. The processing module may periodically poll various peripheral devices and interfaces, so that the computing device can input and output data through various peripheral devices and interfaces, and can also communicate with another computing device having the hardware elements described above.

[0044] The following first uses " Figure 1 "The schematic diagram of the components of the device for analyzing boxing movements to provide boxing training suggestions proposed in the present invention is used to illustrate the device for implementing the present invention. Figure 1 ” As shown, the device 100 of the present invention includes a memory 110, a camera 120, a communication interface 130, a storage medium 140, an input / output module 150, a processor 170, and a bus 190. The processor 170 is connected to the memory 110, the camera 120, the communication interface 130, the storage medium 140, and the input / output module 150 through the bus 190.

[0045] Memory 110 may store one or more sets of computer instructions.

[0046] The camera 120 may include a circuit board, a lens assembly, and an image sensing element (all not shown in the figure), and the lens assembly and the image sensing element are connected through the circuit board. The camera 120 can obtain images through the lens assembly and the image sensing element. In the present invention, the device 100 is not limited to including one camera 120, and may also include multiple cameras 120 that synchronously obtain images.

[0047] The communication interface 130 can be connected to an external network storage device or a server or other network device, and request and download data from the connected network device.

[0048] The storage medium 140 may store data or signals downloaded by the communication interface 130 , may store data or signals provided to the processor 170 or required by the processor 170 for operation, and may store data or signals generated by the processor 170 .

[0049] The input / output unit 150 may provide input data through peripheral input devices of the device 100. For example, the input / output unit 150 may input data through a keyboard, a mouse, a touch pad, or a touch screen.

[0050] The input / output unit 150 may also output the data generated by the processor 170 through a peripheral output device of the device 100. For example, the input / output unit 150 may display the data through a display or a touch screen.

[0051] The processor 170 may be configured as follows: Figure 2 "As shown in the module diagram of the present invention, it includes modules such as data acquisition module 210, data analysis module 230, opponent simulation module 250, reality interaction module 270, and suggestion generation module 290. In some embodiments, the processor 170 can execute the computer instructions stored in the memory 110, and can generate " Figure 2 " In each module; in another embodiment of the present invention, " Figure 2 Each module in " may be generated by one or more circuits and / or complete or partial chips and other hardware components, that is, the processor 170 includes components " Figure 2 " The hardware components of each module in the processor 170, that is, each module included in the processor 170 can be a software module or a hardware module, and the present invention has no special limitation.

[0052] The data acquisition module 210 is responsible for acquiring historical game images, and the historical game images acquired by the data acquisition module 210 include the opponents. The data acquisition module 210 can read the historical game images from the storage medium 140 or acquire the historical game images received by the communication interface 130 .

[0053] The data analysis module 230 is responsible for analyzing the historical game images obtained by the data acquisition module 210 to generate opponent action information of the game opponent. The opponent action information generated by the data analysis module 230 includes the opponent's punching action or defensive action, position, footwork and movement speed. In some embodiments, the opponent action information may also include the opponent's punching force, and the opponent's fatigue and physical condition analyzed based on the punching force.

[0054] For example, the data analysis module 230 can analyze the opponent's movements and steps through computer vision technology and deep learning models, such as posture estimation, skeleton detection and action classification, to generate opponent action information, including punching methods (such as straight punches, hooks, uppercuts, etc.), punching frequency, punching force, punching angle, punching speed, dodge speed, dodge direction, blocking position, etc. At the same time, the data analysis module 230 can also attach the analyzed opponent action information to the historical game images.

[0055] The opponent simulation module 250 is responsible for training the machine learning model based on the opponent's action information and historical game images.

[0056] The opponent simulation module 250 is also responsible for using the trained machine learning model to generate action strategies and action patterns of a virtual opponent that simulates a competition opponent. The action strategies generated by the opponent simulation module 250 include offense or defense, footwork, and movement speed. The action patterns generated by the opponent simulation module 250 correspond to the action strategies. When the action strategies include offense, the action patterns include punching frequency, method, speed, strength, and footwork; and when the action strategies include defense, the action patterns include blocking or evasive positions, and footwork.

[0057] For example, the opponent simulation module 250 can use a machine learning model (such as a decision tree or a reinforcement learning model) to simulate the action pattern of the opponent. Through the analysis of the game data, the machine learning model can learn the behavior patterns of the opponent in different situations. For example, the preferred actions and coping strategies of the opponent in attack and defense can be learned based on past game data, and the timing of the opponent's attack, the situation of defense or counterattack, the offensive / defensive / counterattack actions, the footwork of controlling the position on the field, etc. can be judged accordingly.

[0058] The opponent simulation module 250 can also use a deep learning model, such as a long short-term memory network (LSTM) or a recurrent neural network (RNN) to learn and predict the user's action sequence, and control the virtual opponent to defend or dodge according to the user's fighting actions and the current defense mode. For example, it can make instant responses to each punch or movement of the user and adjust the tactical strategy accordingly, such as how to evade, counterattack or attack, etc. The opponent simulation module 250 can also generate movement decisions for the virtual opponent based on the position and pace of the virtual opponent.

[0059] The opponent simulation module 250 can also use the action data and tactical data to perform pattern recognition, classify the user's behavior into different types, such as offensive, counterattack, or defensive, and determine the game strategy based on the classified type, such as focusing on defense or fast attack, etc., and can dynamically adjust the strategy according to the progress of the game and the user's behavior, simulate the actions taken by the opponent in different situations, and can use one or more of the above-mentioned deep learning models to predict the user's actions based on the user's actions, and make corresponding responses, for example, predict the user's next attack direction based on the user's punching pattern, and defend or counterattack in time, so as to realistically simulate the instant reaction ability of human opponents. Among them, the opponent simulation module 250 can also use the reinforcement learning (RL) model to try to simulate a combination of multiple punching actions and defensive actions, and adjust the action strategy and action pattern according to the simulation results, and the above-mentioned simulation results include the number of times or scores of the user hitting the opponent, the number of times the user is hit, etc.

[0060] The reality interaction module 270 can establish a practice field for the user to practice fighting with the virtual opponent, so that the user and the virtual opponent can practice fighting in the practice field. The reality interaction module 270 can load the pre-set field data to establish the practice field, or obtain the corresponding field data according to the pre-set field conditions to establish the practice field, and the present invention has no particular limitation. Among them, the reality interaction module 270 can read the field data from the storage medium 140 or obtain the field data through the communication interface 130.

[0061] The reality interaction module 270 is responsible for providing the user with fighting movements to practice fighting with the virtual opponent through the reality technology according to the action strategy and action pattern generated by the opponent simulation module 250. For example, the reality interaction module 270 can obtain the action and footwork data of the opponent from the database, including common action types, footwork patterns, attack habits and defense strategies, and can generate smooth action transitions through motion capture technology. For example, when the virtual opponent transitions from a defensive posture to an offensive posture, it is ensured that the virtual opponent's actions are natural and coherent to avoid sudden action changes.

[0062] The reality interaction module 270 is also responsible for obtaining fighting action data corresponding to the fighting action of the user during fighting practice. Generally speaking, the reality interaction module 270 can obtain the fighting action data through the data acquisition module 210, for example, by using motion capture technology to analyze the fighting action of the user to obtain the fighting action data, but the present invention is not limited thereto.

[0063] The reality interaction module 270 also generates the user's target action information. Generally speaking, the reality interaction module 270 can analyze the obtained fighting action data through the data analysis module 230 to generate the target action information. In the present invention, the target action information includes attack information, defense information, and movement information. The attack information includes punching angle, punching rhythm, and successful hit rate; the defense information includes the height and position of the block, the effect of the block, the success rate of the block, the timing of the dodge, the number of dodges, and the success rate of the dodge; the movement information includes the body's center of gravity and posture, and the coordination of footwork.

[0064] The suggestion generation module 290 is responsible for generating training suggestions corresponding to the opponent according to the target action information. The training suggestions generated by the suggestion generation module 290 may include defensive strategies, attack methods or footwork movements for the opponent, and may also include action improvement suggestions, game strategy suggestions, simulation training plans, etc. for the user.

[0065] For example, the suggestion generation module 290 can analyze whether the user's punching angle is appropriate and whether it is easy to be predicted by the opponent, and provide specific suggestions such as adjusting the angle or speeding up the punching speed. It can also detect that the user's punching rhythm or continuity is poor, and suggest how to adjust the number and interval of combos to enhance the fluency and unpredictability of the attack. It can also adjust the punching strategy when the opponent often counterattacks. The suggestion generation module 290 can also analyze the user's defensive movements, improve the timing of evasive movements, or adjust the height and position of blocking movements. It can also suggest how the user adjusts the center of gravity and posture of the body to ensure that the user can flexibly respond to attacks while maintaining his own defensive balance. The suggestion generation module 290 can also generate footwork improvement suggestions when it detects that the user's footwork and punching are not coordinated, so as to help the user improve the continuity of movement and punching and avoid slow movement or unstable rhythm. It can also suggest that the user speed up or slow down his footwork and maintain stability during movement to better control the rhythm of the game. It can also adjust the movement footwork when the opponent often counterattacks.

[0066] Next, an embodiment is used to explain the system operation and method of the present invention, and please refer to " Figure 3 ” A flow chart of the method for analyzing boxing movements to provide boxing training suggestions according to the present invention. In this embodiment, it is assumed that the device 100 is a server providing mixed reality services, but the present invention is not limited thereto.

[0067] First, the data acquisition module 210 of the device 100 can acquire historical game images including the opponents (step 310). In this embodiment, it is assumed that the administrator of the server can store a large number of historical game images in the storage medium 140 of the device 100 in advance, and the data acquisition module 210 can read the historical game images from the storage medium 140. The administrator can also input the network address or domain name of the network storage device storing the historical game images and the storage path of the historical game images through the input / output module 150 of the device 100. The data acquisition module 210 can connect to the network storage device through the communication interface 130 of the device 100 according to the network address or domain name and the storage path to read the historical game images.

[0068] After the data acquisition module 210 of the device 100 acquires the historical game video (step 310), the data analysis module 230 of the device 100 can analyze the historical game video to generate the opponent's opponent's opponent's opponent's opponent action information (step 320). In this embodiment, it is assumed that the data analysis module 230 can analyze the opponent's opponent's punching method, punching frequency, punching force, punching angle, punching speed, dodge speed, dodge direction, blocking position and other action data when analyzing the historical game video, and can also analyze the opponent's punching force, and can judge the opponent's opponent's fatigue and physical condition based on the analyzed punching force.

[0069] After the data analysis module 230 of the device 100 generates the opponent action information of the game opponent, the opponent simulation module 250 of the device 100 can train the machine learning model based on the historical game images obtained by the data acquisition module 210 and the opponent action information generated by the data analysis module 230 of the device 100 (step 330).

[0070] After the opponent simulation module 250 of the device 100 completes the training of the machine learning model, the communication interface 130 of the device 100 can provide the user with a connection to the real device 400. After the real device 400 is connected, the opponent simulation module 250 can use the machine learning model to generate the action strategy and action pattern of the virtual opponent that simulates the opponent (step 340). In this embodiment, it is assumed that the opponent simulation module 250 can provide an opponent selection interface to the real device 400 through the communication interface 130, so that the user can operate the real device 400 to select the opponent in the opponent selection interface, and can obtain the opponent selected by the user through the real device 400 through the communication interface 130, and use the machine learning model to generate the corresponding action strategy and action pattern based on the database of the selected opponent, wherein the opponent simulation module 250 can try to simulate a combination of multiple punching actions and defensive actions, and adjust the action strategy and action pattern according to the simulation results, so that the adjusted action strategy and action pattern include the most suitable combination of punching actions and defensive actions.

[0071] After the opponent simulation module 250 of the device 100 generates the action strategy and action pattern (step 340), the reality interaction module 270 of the device 100 can provide the user with fighting actions to practice fighting with the virtual opponent through the reality technology according to the generated action strategy and action pattern (step 350), and can obtain the fighting action data corresponding to the fighting actions during the fighting practice between the user and the virtual opponent (step 360). In this embodiment, it is assumed that the reality interaction module 270 can first establish a practice field according to the pre-set field conditions, and can also establish a virtual opponent (if the reality technology provided by the reality interaction module 270 is virtual reality, a virtual character operated by the user can be established at the same time), and can provide the user with the virtual opponent to practice fighting in the practice field through the reality device 400, for example, the action strategy and action pattern of the virtual opponent are transmitted to the reality device 400, so that the reality device 400 displays the actions of the virtual opponent to the user according to the received action strategy and action pattern, so as to practice fighting with the user, wherein the reality interaction module 270 can In order to obtain the fighting movements of the user against the virtual opponent in the training ground, and generate corresponding fighting movement data based on the obtained fighting movements. At the same time, the reality interaction module 270 can also select fighting interaction information that responds to the obtained fighting movement data based on the action strategy and action mode, and control the virtual opponent to perform corresponding fighting movements based on the selected fighting interaction information, so as to instantly control the virtual opponent to fight against the user. For example, according to the fighting movement data and the current defense mode determined by the action strategy and action mode, the virtual opponent is controlled to defend against the user's fist or dodge the user's fist, and the movement decision of the virtual opponent is continuously generated according to the position and pace of the virtual opponent.

[0072] After the real-world interaction module 270 of the device 100 obtains the fighting action data of the user against the virtual opponent (step 360), the data analysis module 230 of the device 100 can analyze the fighting action data to obtain the user's target action information (step 370), and the suggestion generation module 290 of the device 100 can generate training suggestions corresponding to the opponent based on the target action information generated by the data analysis module 230 (step 380).

[0073] In summary, it can be seen that the difference between the present invention and the prior art lies in the technical means of analyzing the historical game images of the opponent to generate the opponent action information of the opponent, and using the machine learning model trained based on the opponent action information of the opponent to simulate the action strategy and action pattern of the opponent, providing the user with a virtual opponent to practice fighting with the virtual opponent through reality technology based on the action strategy and action pattern, and analyzing the user's fighting movements in the fighting practice to generate target action information, and generating training suggestions corresponding to the opponent based on the target action information. This technical means can solve the problem that the boxing training system in the prior art cannot respond immediately according to the user's actions and cannot give the user improvement suggestions based on the user's performance, thereby achieving the technical effect of truly simulating the opponent's reaction and giving overall improvement suggestions.

[0074] Furthermore, the method of analyzing boxing movements to provide boxing training suggestions of the present invention can be implemented in hardware, software, or a combination of hardware and software and can be implemented in a centralized manner in a computer system or in a distributed manner where different components are spread across several interconnected computer systems.

[0075] Although the embodiments disclosed in the present invention are as above, the above contents are not used to directly limit the scope of patent protection of the present invention. Any technician in the technical field to which the present invention belongs, without departing from the spirit and scope disclosed in the present invention, makes slight changes and modifications to the form and details of the implementation of the present invention, which are within the scope of patent protection of the present invention. The scope of patent protection of the present invention shall still be based on the scope defined by the attached claims.

Claims

1. A method for analyzing boxing movements to provide boxing training suggestions is applied to a device or system, the method comprising at least the following steps: Obtaining historical match images, which include match opponents; Analyzing the historical match video to generate opponent action information of the opponent, the opponent action information including punching or defensive action, position, footwork and moving speed; Training a machine learning model based on the historical game images and the opponent's action information; Using the machine learning model to generate action strategies and movement patterns of a virtual opponent simulating the opponent, the action strategies include attack or defense, footwork, and movement speed, and the movement patterns include punching frequency, method, speed, strength, or blocking position; According to the action strategy and the action mode, the user is provided with fighting actions to practice fighting with the virtual opponent through real-life technology; Obtaining fighting action data corresponding to the fighting action; Analyzing the fighting action data to obtain target action information of the user, the target action information including attack information, defense information, and movement information; and A training suggestion corresponding to the opponent is generated according to the target action information.

2. The method for analyzing boxing movements to provide boxing training suggestions as described in claim 1, wherein the step of analyzing the historical game images to obtain the opponent's action information also includes obtaining the opponent's punching force and determining the opponent's fatigue and physical condition.

3. The method for analyzing boxing movements to provide boxing training suggestions as described in claim 1, wherein the step of using the machine learning model to generate the action strategy and the action pattern of the virtual opponent that simulates the opponent is to try to simulate a combination of multiple punching movements and defensive movements, and adjust the action strategy and the action pattern according to the simulation results.

4. The method for analyzing boxing movements to provide boxing training suggestions as claimed in claim 1, wherein the step of providing the user with the virtual opponent with the action strategy and the action pattern through real-life technology to perform the sparring practice further comprises the step of establishing a practice field for performing the sparring practice based on pre-set field conditions.

5. The method for analyzing boxing movements to provide boxing training suggestions as claimed in claim 1, wherein the step of providing the user with the fighting movements to perform the fighting practice with the virtual opponent through reality technology based on the action strategy and the action pattern further comprises the step of controlling the virtual opponent to defend or dodge according to the fighting movement data and the current defense mode, and generating a movement decision for the virtual opponent according to the position and pace of the virtual opponent.

6. A device for analyzing boxing movements to provide boxing training suggestions, the device comprising at least: A data acquisition module is used to acquire historical game images, where the historical game images include game opponents; A data analysis module, for analyzing the historical game images to generate opponent action information of the game opponent, the opponent action information including punching or defensive action, position, footwork and moving speed; An opponent simulation module, for training a machine learning model based on the historical game images and the opponent's action information, and for using the machine learning model to generate an action strategy and action pattern of a virtual opponent simulating the game opponent, wherein the action strategy includes attack or defense, footwork, and movement speed, and the action pattern includes punching frequency, method, speed, strength, or blocking position; A reality interaction module is used to provide the user with fighting movements to practice fighting with the virtual opponent through reality technology according to the action strategy and the action mode, so that the data acquisition module obtains fighting movement data corresponding to the fighting movement, and the data analysis module is also used to analyze the fighting movement data to obtain the user's target movement information, and the target movement information includes attack information, defense information, and movement information; and The suggestion generating module is used to generate training suggestions corresponding to the competition opponent according to the target action information.

7. The device for analyzing boxing movements to provide boxing training suggestions as claimed in claim 6, wherein the opponent action information further includes the punching force of the opponent, the fatigue level and physical condition of the opponent.

8. The device for analyzing boxing movements to provide boxing training suggestions as claimed in claim 6, wherein the opponent simulation module is further used to try to simulate a combination of multiple punching movements and defensive movements, and adjust the action strategy and the action pattern according to the simulation results.

9. The device for analyzing boxing movements to provide boxing training suggestions as claimed in claim 6, wherein the reality interaction module is further used to establish a practice field for performing the sparring practice according to pre-set field conditions, so that the user and the virtual opponent can perform the sparring practice in the practice field.

10. The device for analyzing boxing movements to provide boxing training suggestions as claimed in claim 6, wherein the opponent simulation module is further used to control the virtual opponent to defend or dodge according to the fighting movement data and the current defense mode, and to generate a movement decision of the virtual opponent according to the position and pace of the virtual opponent.

Images