470 results about "Sport training" patented technology

The invention discloses a three-dimensional attitude high-precision acquisition and reconstruction system for exercise training, which relates to the technical field of three-dimensional attitude acquisition and reconstruction and comprises a camera calibration module, an image acquisition module, a key point extraction module, a three-dimensional reconstruction module, an attitude optimization module, a data fusion module and a feedback generation module. The camera calibration module is used for performing internal and external parameter calibration on a multi-view camera assembly by adopting a self-adaptive joint calibration method to obtain a projection matrix parameter and a transformation matrix of each camera; the image acquisition module is used for synchronously acquiring multi-view video images of the athlete in the training process based on the calibration parameters, and performing timestamp alignment processing to obtain an original image sequence; and the key point extraction module is used for detecting and identifying a human body image by adopting a deep learning model, extracting two-dimensional coordinates of key human body joint points of each frame from an original image sequence, and obtaining a two-dimensional key point set under each view angle.

The invention provides a physical training posture correction method based on machine vision, which comprises the following steps: acquiring an athlete training image sequence through a multi-view image acquisition system and preprocessing the athlete training image sequence, extracting three-dimensional posture key points of a human body by utilizing a depth posture recognition model, constructing a skeleton model and calculating real-time kinematics characteristic parameters, and performing multi-dimensional comparison with standard parameters to generate a posture deviation evaluation report, thereby generating a personalized multi-modal correction feedback scheme, and establishing a personal athlete movement feature database to realize adaptive standard parameter optimization. According to the method, the training postures of athletes can be accurately recognized and corrected in real time, the training effect can be improved, sports injuries can be prevented, and training individuation and scientificity are improved.

The invention is suitable for the technical field of 3D scanning, and provides a figure model 3D scanning and dynamic attitude capturing method and system, which dynamically triggers 3D scanning by monitoring joint movement intensity in real time, continuously collects continuous movement tracks by using an inertial sensor in a non-scanning period, and reversely deduces space-time parameters of a shielded part through multi-light-source shadow boundary evolution. And constructing a spatio-temporal joint evaluation function to adaptively adjust the scanning frequency and the sampling rate, and finally performing spatio-temporal alignment and confidence weighted fusion on the discrete point cloud, the continuous trajectory and the reverse reckoning data to generate a dynamic biomechanical model. The scheme breaks through the bottleneck of resource waste and blind area shielding of traditional fixed frequency scanning, realizes space-time complementation and precision optimization of motion monitoring, and is suitable for dynamic scenes such as movie and television animation, medical rehabilitation, physical training and the like. According to the system, through multi-modal data fusion and self-adaptive resource scheduling, the data integrity, the real-time performance and the biomechanical analysis capability are remarkably improved, and an efficient solution is provided for human motion digitization.

The invention relates to the technical field of intelligent fitness equipment and exercise training control, and provides a training plan dynamic generation and action quality scoring method and system for an intelligent dumbbell. The method comprises the following steps: acquiring action data of a user during training and personalized parameters input by the user in real time through an acceleration sensor, an angular velocity sensor and a pressure sensor which are arranged in the intelligent dumbbell; inputting the action data and the personalized parameters into a preset training plan generation algorithm module so as to generate a stage training plan comprising the action type, each group of training duration, intermittent time and a weight adjustment strategy; an action quality scoring model is constructed, a standard action template is pre-stored in the scoring model, and the standard action template comprises standard trajectory curves, speed threshold ranges, amplitude reference values and force exerting symmetry parameters corresponding to different training actions; and performing multi-dimensional matching analysis on the action data and a standard action template, and dynamically correcting the score in combination with the current fatigue parameter of the user.

The invention relates to the technical field of sailboat training instruments, and provides a real-time simulation platform for sailboat exercise training, which comprises a training sailboat used for trainee driving training; the rainwater simulation mechanism is used for simulating the influence of raining on the sight of the driver in rainy days; the water inlet mechanism is used for simulating the influence on the operation of a driver when the training sailboat inclines in rainy days and water enters a cockpit; and the water pumping mechanism is used for simulating internal drainage of the cockpit and recycling recycled water resources. Through the cooperation of the rainwater simulation mechanism, the control mechanism and the water inlet mechanism, the sailboat training simulation platform can simulate the influence of rainwater blowing to a cab from different directions on the sight of a student, can simulate the influence condition of the amount of different rainwater blowing to the interior of the cab from different directions on the sight of the student, and meanwhile, can simulate the influence condition of the amount of rainwater blowing to the interior of the cab from different directions. The sailboat training simulation platform can simulate the influence of rainwater on the deck of the training sailboat entering a cockpit due to the inclination of the boat body on the operation and driving of trainees in rainy days.

The invention discloses a badminton player action posture analysis system and method based on multi-sensor fusion, and relates to the technical field of athletic training auxiliary systems.The system comprises a wearable sensing subsystem, an environment sensing subsystem and a central processing and feedback subsystem; the wearable sensing subsystem is arranged on an inertia measurement unit and a pressure sensing insole of a body to collect movement inertia and plantar pressure data; the environment perception subsystem collects global videos through multiple cameras. The central processing and feedback subsystem performs synchronous processing on multi-source data, adopts a hierarchical fusion algorithm, restrains inertia integral drift by utilizing a plantar contact state, reconstructs a three-dimensional skeleton posture and a motion trail in combination with visual key points and skeleton restraint, drives a digital twin model, compares with a standard motion model, and performs three-dimensional motion control on the three-dimensional skeleton posture and the motion trail. Quantitative evaluation results such as joint angle deviation, time sequence difference and pressure center track are generated, and visual feedback is performed through a display device and an augmented reality terminal for badminton training evaluation and technical deviation correction.

A system for optimizing mechanics and movements in remote or outpatient physical therapy, rehabilitation, sports training, and injury prevention that uses several inertia measurement units (IMUs) to measure a user's motion while performing an action. The IMUs can have additional sensors connected to improve the system's ability to detect flaws in the user's motion. Furthermore, the system uses machine learning to detect and determine flaw in a user's motion from the IMU data. The system may generate feedback to improve the user's motion based on the detected flaws. Different feedback communication may be provided based on the performance of the user after the feedback is provided.

This disclosure pertains to a sports training system that utilizes deep learning and computer vision technologies to analyze biomechanical sport motions. The system is equipped with AI-powered functionalities to enhance training, including those related to creating and managing motion templates that integrate video sequences with extracted motion features. These templates serve as benchmarks, accessible to trainees for improving their technique and performance. The system also can employ an advanced motion synchronization process, which combines dynamic time warping and computer vision technologies to align trainee videos with reference motions captured in motion templates. Additionally, a motion feature database can be updated with features extracted from both trainee and trainer videos, storing visual information in a structured format to facilitate various analytic functions. The system supports analysis of a wide range of sports motions.

The present invention discloses an information-based sports education platform management system, which specifically relates to the field of sports training data management, and is used to solve the problem of accurate classification and feedback of multi-device training data. Firstly, based on the fluctuation characteristics of physiological data, the switching period data is captured to ensure the continuity and timeliness of the training data; then, the transition response consistency index and the target device feature matching degree are used to accurately define the effective transition data, clearly distinguish the physiological feature data of each device, and improve the classification accuracy; the defined effective transition data is finely classified and labeled through the target device dynamic feature model to ensure the independence and consistency of the data; finally, an adaptive rejection mechanism is used to screen out abnormal data, retain high-quality data to generate personalized feedback, enhance the pertinence of the feedback, and build a closed-loop system from data collection to feedback generation, providing scientific support for sports training data management.

The invention discloses a basketball player posture detection method and system based on deep learning, relates to the technical field of sports science, and is used for solving the problem of accuracy of player posture detection in basketball sports. According to the method, basketball motion videos are collected from multiple visual angles, the videos are preprocessed, video quality scores are calculated, and the videos are classified. Different key point optimization detection modes are adopted for A-class videos and B-class videos, further, myoelectricity and electroencephalogram signals are fused, the physiological state score of an athlete is calculated, the attitude key point confidence coefficient is dynamically adjusted, and the detection accuracy is improved. And finally, calculating the matching degree of the attitude key point sequence in the video through dynamic time warping. The method can detect the posture of the basketball player at high precision in a complex environment, has high robustness, and is suitable for exercise training and tactical analysis.

A method for providing visual guidance for sports training is disclosed. The method advantageously captures and analyzes the real-time motions of a novice athlete and compares them with the previously recorded motions of an expert athlete. Additionally, the method advantageously provides both embodied and detached visual guidance for the novice athlete and, in particular, provides feedback that helps the novice athlete to quickly and intuitively understand the differences between their real-time motions and the previously recorded motions of the expert athlete.

The invention relates to the technical field of wearable intelligent equipment, and further relates to an exercise training plan generation method and system and wearable intelligent equipment. The method comprises the following steps: when a user is not in a motion state, determining a first maximum heart rate and a first maximum oxygen uptake of the user based on a support vector regression model and basic information of the user; when the user is in the motion state, photoelectric volume pulse wave signals generated by the user in the motion process are processed through a deep learning model, and a plurality of basic index sequences are obtained; determining a second maximum heart rate and a second maximum oxygen uptake of the user based on the plurality of basic index sequences, the basic information, the maximum heart rate and a maximum oxygen uptake prediction model; and generating an exercise training plan of the user based on the first maximum heart rate and the first maximum oxygen uptake, or generating an exercise training plan of the user based on the second maximum heart rate and the second maximum oxygen uptake. The method reduces the generation cost of the exercise training plan.

The invention discloses a training method and device for controlling an intelligent mouse, equipment, a storage medium and a program product, and relates to the technical field of game operation training. In the invention, a new game operation training method is provided, mouse operation content of a top player in a target game segment is recorded in advance, and then the target game segment is trained according to a training template composed of a game scene of the target game segment and the corresponding mouse operation content. The method comprises the steps that a target game segment is obtained, a game scene of the target game segment is reconstructed, mouse operation content is prompted, mouse operation triggered by a to-be-trained person and a learner of the game operation under the prompt of the mouse operation content is received, and next-step training guidance corresponding to the current mouse operation is further fed back. Compared with traditional electronic sports training, long-term repeated training does not need to be performed depending on a large number of random game scenes, and the target game operation in the to-be-trained target game scene is directly trained, so that the game operation training efficiency and training effect are improved.

The invention discloses a memory enhancement action recognition method and system on a Riemannian manifold and a storage medium, and the method comprises the steps: 1, collecting human body action data, and representing the human body action data as a third-order tensor; 2, expanding along three modes to obtain three corresponding matrixes; 3, calculating by adopting a human short-term memory mechanism to obtain a memory enhanced weight matrix; 4, decomposing the weight matrix through a principal component analysis method to obtain a base vector with a weight; 5, recombining, normalizing and mapping to a unit hyper-sphere, and reserving an angle relation; 6, learning modal weight parameters through a Monte Carlo Markov algorithm, and calculating geometric differences between points on the hypersphere; and 7, carrying out human body action classification by adopting a K-nearest neighbor classifier. The method effectively solves the problem of time information loss in a complex scene, and is suitable for various application scenes such as medical health, virtual reality, physical training and the like.

A sports training system can include a target board and a controller configured to control operations of the target board. The target board can be configured to be hit by a ball that is struck by a player, and rebound the ball to the player. The target board can include a plurality of board zones and vibration sensors respectively corresponding to the board zones. Each vibration sensor can sense vibration of a respective board zone. The controller can control illumination of a light source to designate one of the board zones as a target board zone for the player to hit with the ball, and determine whether the ball has hit the target board zone based on whether a vibration signal characteristic of the target board zone being hit by the ball is received from the vibration sensor corresponding to the target board zone.

The invention discloses a physical training posture collection method and system based on a model database and an adjustment method, and belongs to the technical field of physical training, and the method comprises the following steps: S1, transmitting a single echo through a laser radar, and collecting human body point cloud data and image information in combination with a high-definition camera; s2, performing multi-dimensional maximum pooling operation on the point cloud data based on PointNet to generate a three-dimensional human body model; s3, extracting human body key points by using an RMPE technology, and combining the human body key points with the SMPL parameterized model; and S4, segmenting a multi-person view through Mask R-CNN, and carrying out comparative analysis on real-time action data and a champion model database. According to the physical training posture collection method and system based on the model database and the adjustment method, motion data are collected through a laser radar and a high-definition camera, a three-dimensional human body model is constructed by utilizing a point cloud processing technology, posture comparison and analysis are performed in combination with the champion model database, and low-delay feedback is realized through a real-time operation system.

The invention discloses an athlete action analysis and training optimization method and system based on big data, and relates to the technical field of exercise training.Multiple array sensors are deployed to collect exercise data, including static data and dynamic data, of a trainee, and meanwhile medical images of the trainee are obtained; constructing a basic model of the trainer based on the motion data, wherein the basic model comprises a body model and a motion model; acquiring athlete big data to construct an athlete big database, and generating a shadow model based on coupling of the athlete big database and the body model; calculating the coincidence degree of the basic model and the shadow model, generating an optimal function unit based on the coincidence degree, and guiding training of other body function units based on the optimal function unit to obtain training data. According to the method, the trainee basic model and the athlete shadow model are constructed, a scientific target reference is set for personalized training, and the pertinence and operability of training optimization are greatly improved through data-driven personalized model construction.

The invention relates to a physical training scheme generation method and system for limb feature recognition, and relates to the field of physical training, and the method comprises the steps: collecting a skeleton topological portrait and a body shape proportion portrait of a target user; retrieving a joint number list and a joint standard time sequence coordinate list of the to-be-trained motion type meeting the portrait; when the target user is trained, the joint training time sequence coordinate list is extracted and compared with the joint standard time sequence coordinate list to obtain a joint time sequence deviation coordinate list; traversing the joint time sequence deviation coordinate list to count a time proportion list of which the coordinate deviation distance is greater than or equal to a deviation distance threshold value; the joint number set with the moment proportion larger than or equal to the moment proportion threshold value is extracted, the joint number set is input into the action matching library to execute abnormal action configuration to obtain an abnormal action set, a physical training scheme is configured, and the technical problems that due to the fact that physical training neglects individual body difference, monitoring precision is insufficient, and the individualization degree of the physical training scheme is poor are solved.

The invention relates to a static balance real-time control method and system for a wearable tactile feedback device, and aims to provide an efficient, accurate and real-time static balance control scheme. According to the system, COP position data of a user is collected in real time through the pressure center COP collection module, and timely normal pressure feedback is provided in combination with the wearable tactile feedback device, so that the balance state of the user is adjusted. The system adopts an advanced data processing algorithm, supports multiple feedback control mode linear, quadratic and logarithmic functions, and can calculate COP offset in real time and generate corresponding feedback signals. Through the interactive interface, a user can conveniently set control parameters and display the COP track in real time. The system is widely applied to the fields of exercise training, rehabilitation treatment and virtual reality, has high-precision and low-delay feedback control capability, can effectively improve the accuracy and real-time performance of balance adjustment, fills the blank in the prior art, and has relatively strong application prospects and market potential.

The invention discloses a personalized physical training load intelligent regulation and control system based on multi-modal physiological data fusion, and relates to the technical field of intelligent physical training and health monitoring. The system comprises a multi-modal data acquisition module, an intelligent fusion analysis module, a personalized decision module and a real-time regulation and control module. The method comprises the following steps: acquiring physiological, motion and environmental parameters of a user, and performing multi-modal data fusion by using a time sequence-causal attention mechanism to generate a comprehensive physiological state feature vector; a dynamic personalized baseline is established based on a meta-learning framework, a load adjustment decision is generated by adopting a hierarchical reinforcement learning algorithm, and regulation and control safety is ensured through causal reasoning verification; and finally, load adjustment is executed through the real-time regulation and control module to form closed-loop control. According to the method, the problems that in the prior art, training load regulation depends on a single physiological parameter, the individuation degree is insufficient, and decision making is lack of scientificity are solved, and accurate, self-adaptive and safe individualized physical training management is achieved.

The invention relates to the field of medical care informatics, and discloses an exercise training management system based on multi-sensor fusion, which is characterized in that the system acquires calibration data to establish an individualized exercise physiological response baseline model, inputs actual exercise performance into the baseline model in training, and determines expected physiological signs; calculating a time sequence deviation value between the expected physiological sign and the actual physiological sign; when it is monitored that the athletic performance declines, attribution is carried out based on a time sequence deviation value, and the athletic performance declines are judged to be derived from conventional fatigue or potential physiological health risks, by introducing a time sequence deviation attribution mechanism, the technical problem that in the prior art, due to data isolation, the athletic performance decline causes cannot be judged is solved, and the probability that the athletic performance declines is lowered is lowered. Therefore, the method has the capability of distinguishing conventional fatigue from potential health risks, and decision conflicts between invalid exercise guidance and health risk early warning are avoided.

The invention relates to a sports competition training action deviation detection and immediate correction method. The method comprises the following steps: collecting multi-joint dynamic data in a training process; on the basis of a preset inertial conduction continuity criterion, inertial change among all joints of the body is analyzed, and when it is detected that inertial conduction of any joint node is interrupted or an impact value is lower than a threshold value, the joint node is identified as a trigger point of action chain interruption; performing causal priority judgment on the action deviation based on the identified trigger point, determining whether the deviation is a dominant deviation or a subordinate deviation caused by an upstream joint, and if the deviation is the subordinate deviation, tracing an upstream cause and repositioning an intervention target; judging whether the current training stage is in a correction window stage suitable for feedback or not for the deviation which is confirmed to be intervened, and if not, delaying feedback output; and in the window period, trend judgment is performed on the deviation direction, if the deviation causes negative technical degradation, correction feedback is output, and if the deviation trend has positive evolution potential, the deviation is marked as acceptable deviation and is not corrected.

A pressure-sensitive stacking structure for a sports training device of the present invention that is able to detect a pressure with which a player grips a grip portion may include: a conductive layer that is pressure-sensitive; a circuit layer disposed to face the conductive layer and having a plurality of conductive patterns; and a separating layer disposed between the conductive layer and the circuit layer and formed of a fibrous tissue including a plurality of fibers, wherein, when a pressure is generated, the conductive layer and the circuit layer may come into contact, and the pressure may be detected.

The present invention discloses a risk warning method for sports training injuries of athletes, which belongs to the field of sports risk warning technology and includes the following steps: S1: collecting limb movement information and limb movement trajectory of athletes in a sports training state, and determining the sports training information of the athletes in the sports training state; S2: pre-processing the collected sports training information of the athletes in the sports training state, and determining sports training characterization data of the athletes in the sports training state; S3: analyzing and evaluating the sports training characterization data of the athletes in the sports training state. The present invention solves the existing problem that it is impossible to provide timely risk warnings for sports training injuries of athletes, resulting in poor sports training effects of athletes. The present invention can provide timely risk warnings for sports training injuries of athletes, avoid injuries to athletes during sports training, facilitate the prevention of sports training injuries of athletes, and improve the sports training effects of athletes.

The invention relates to the technical field of training teaching, in particular to an exercise training action teaching system. The system comprises a training data acquisition module, a trainee action difference analysis module, a trainee training teaching module and a demonstration teaching feedback adjustment module, and can acquire corresponding training action postures and acceleration data in real time by configuring an inertial sensor on sports equipment corresponding to a trainee; meanwhile, the corresponding body consumption load condition is obtained; performing student action difference feature analysis on the training action postures and the acceleration data based on the body consumption load condition to obtain training action difference features corresponding to different students in the daily exercise training process; training teaching resources and time limiting conditions are obtained, and student training teaching analysis and teaching feedback adjustment are carried out, so that an exercise training teaching feedback adjustment strategy corresponding to the student is generated. According to the invention, scientific and personalized action teaching services can be provided for exercise training learners, so that the exercise training effect is improved.

The invention discloses a multi-target physical training action evaluation feedback system based on artificial intelligence, and the system obtains video data and physiological data in a multi-target motion scene through a multi-dimensional data collection module, and optimizes skeleton key point data through a data preprocessing module; and the continuous identity association of the moving target is realized by the multi-target tracking and binding module. The motion modeling module constructs motion state variables based on a target constraint dynamic optimization model and applies physiological and physical constraints, the fusion evaluation module combines local motion decomposition and physiological data correlation analysis to generate a comprehensive score, and the personalized feedback module outputs hierarchical improvement suggestions. And the adaptive standard library module iteratively optimizes standard action parameters according to the user type and historical data. The system solves the problems that in the prior art, multi-target tracking is confused, action evaluation and physiological states are disjointed, and feedback is lack of accuracy, and efficient analysis, accurate evaluation and personalized guidance of actions in a multi-target scene are achieved.

The present application relates to the field of computer-aided sports training equipment, and particularly relates to a method and device for generating a demonstration video for sports training, computer equipment and a storage medium, wherein a demonstration action is constructed based on a composite action of a user and a decomposition action template corresponding to each decomposition action of the composite action, a more targeted decision basis can be provided for the sports training of the user according to the actual training condition of the user, and customization of the training content is realized.

The invention discloses a safety risk early warning method and system for physical training, and relates to the technical field of physical training. Obtaining a quantum biological field abnormal coefficient, a muon bone mineral density variation coefficient, a motion posture coefficient and a training environment coefficient of each part of each trainee in real time; according to the coefficients, the safety risk value of each part of each trainee is accurately analyzed, and the safety risk evolution trend of each trainee is deeply analyzed; and finally, on the basis of the risk evolution trend, performing optimization and adjustment on protection measures of all parts of all trainees in a targeted manner. According to the method, through multi-dimensional data collection and analysis, dynamic monitoring and early warning of physical training safety risks are achieved, protection measures are optimized in time, the training risks are effectively reduced, reliable guarantee is provided for physical training safety, and the method is suitable for various physical training scenes.

The application relates to the technical field of sports training and discloses an intelligent sensing and feedback personalized sports training guidance system, which comprises a dynamic interactive field sensing module, a data processing and analysis module, an immersive feedback module and a personalized model evolution module; the sensing module collects real-time interactive data through an intelligent training pad integrated with pressure and vibration sensors; the data processing and analysis module is internally provided with a predictive intention engine, which can evaluate a posture before a user action and trigger a feedforward early warning, and calculate an actual motion track of the user after the action, namely a fact path; the personalized model evolution module maintains a baseline model based on historical best performance of the user, and the data processing module calls the model to generate an idealized counterfactual path; the immersive feedback module presents an early warning signal on the training pad and synchronously visualizes playback of the fact path and the counterfactual path. The application realizes high self-adaptation and personalization of training guidance and significantly improves training efficiency.