Method and device for motion analysis of a bobsled start phase

Abstract

The application provides a sledding starting stage motion analysis method and device, wherein the method comprises the following steps: acquiring a video sequence of a target to be analyzed in a sledding starting stage in a sledding process; identifying the target to be analyzed in each video frame in the video sequence and determining motion trajectory information of the target to be analyzed; performing human body key point detection on the target to be analyzed in each video frame to obtain human body key point information of the target to be analyzed in each video frame; and determining a motion analysis result of the target to be analyzed based on the motion trajectory information, the human body key point information and time information. Based on the human body key point detection on the target to be analyzed, the human body key point information of the target to be analyzed is determined, and the motion analysis of the target to be detected is performed based on the obtained data, so that the automatic motion analysis process of the target to be analyzed is realized, the accurate identification and tracking of the motion of the athlete can be realized, the accuracy of the analysis result is improved, and the subjectivity of the human judgment is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of motion analysis, and in particular to a motion analysis method and device for a sled start phase. Background Art

[0002] To scientifically collect athlete parameters and conduct motion analysis in bobsleigh racing, existing methods are generally based on motion capture technology, including infrared motion capture technology, inertial wearable capture technology, and video analysis technology.

[0003] Existing motion capture technology is expensive and has complex operating procedures. Wearable devices can affect bobsled athletes' training and cause signal interference, no video footage, and no displacement indicators, resulting in low accuracy in athlete motion analysis. Summary of the Invention

[0004] The present invention provides a motion analysis method and device for a sled start phase, for improving the accuracy of the analysis of the sled motion process.

[0005] The present invention provides a motion analysis method for a sled start phase, comprising the following steps:

[0006] Obtain a video sequence of the target to be analyzed during the sled departure phase of the sled movement;

[0007] Identifying a target to be analyzed in each video frame in the video sequence, and determining motion trajectory information of the target to be analyzed based on the identified target to be analyzed in each video frame;

[0008] Performing human body key point detection on the target to be analyzed in each video frame to obtain human body key point information of the target to be analyzed in each video frame;

[0009] Based on the motion trajectory information, the human body key point information of the target to be analyzed in each video frame, and the time information of the target to be analyzed during the sled departure phase, motion analysis is performed on the target to be analyzed to determine a motion analysis result of the target to be analyzed.

[0010] According to a motion analysis method for a sled start phase provided by the present invention, obtaining a video sequence of a target to be analyzed during the sled start phase during sled motion includes:

[0011] Synchronously acquiring video data of the target to be analyzed at multiple angles during the sled departure phase based on multiple image acquisition devices and time information of the target to be analyzed during the sled departure phase;

[0012] The video data from the multiple angles are fused to obtain fused video data, and video frames are extracted from the fused video to obtain the video sequence.

[0013] According to a motion analysis method for a sled departure phase provided by the present invention, performing motion analysis on the target to be analyzed based on the motion trajectory information, the key point information of the target to be analyzed in each video frame, and the time information of the target to be analyzed during the sled departure phase, and determining a motion analysis result of the target to be analyzed, including:

[0014] Determining a reaction speed of the target to be analyzed and a sled pushing speed of the target to be analyzed based on the motion trajectory information and the time information;

[0015] Determining motion angle data of the target to be analyzed based on the human body key point information of the target to be analyzed in each video frame, wherein the motion angle data at least includes joint angles, trunk compression angles, and body extension angles;

[0016] A motion analysis result of the target to be analyzed is determined based on the reaction speed, the sled speed, and the motion angle data.

[0017] According to a motion analysis method for a sled start phase provided by the present invention, identifying a target to be analyzed in each video frame in the video sequence includes:

[0018] Based on the target detection algorithm YOLO, the target to be analyzed in each video frame in the video sequence is identified, and the target detection frame of the target to be analyzed in each video frame is determined.

[0019] According to a motion analysis method for a sled start phase provided by the present invention, determining motion trajectory information of the target to be analyzed based on the target to be analyzed identified in each video frame includes:

[0020] Based on the target tracking algorithm Deep SORT, target tracking is performed on the target detection frame in each video frame to determine the trajectory information of the target to be analyzed.

[0021] According to a motion analysis method for a sled start phase provided by the present invention, detecting human key points of the target to be analyzed in each video frame to obtain human key point information of the target to be analyzed in each video frame includes:

[0022] Based on the HRNet key point detection technology, human key point detection is performed on the target to be analyzed in each video frame to obtain the human key point information of the target to be analyzed in each video frame.

[0023] The present invention also provides a motion analysis device for a sled start phase, comprising the following modules:

[0024] A video sequence acquisition module is used to acquire a video sequence of the target to be analyzed during the sled departure phase of the sled movement;

[0025] a trajectory determination module, configured to identify the target to be analyzed in each video frame in the video sequence, and determine motion trajectory information of the target to be analyzed based on the target to be analyzed identified in each video frame;

[0026] A key point determination module is used to detect the human key points of the target to be analyzed in each video frame to obtain the human key point information of the target to be analyzed in each video frame;

[0027] The motion analysis module is configured to perform motion analysis on the target to be analyzed based on the motion trajectory information, the key point information of the target to be analyzed in each video frame, and the time information of the target to be analyzed during the sled departure phase, and determine a motion analysis result of the target to be analyzed.

[0028] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the motion analysis method for the start phase of a sled as described above is implemented.

[0029] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the motion analysis method for the sled start phase as described above is implemented.

[0030] The present invention also provides a computer program product, comprising a computer program, which, when executed by a processor, implements any of the above-mentioned motion analysis methods for the sled start phase.

[0031] The motion analysis method and device for the bobsleigh start phase provided by the present invention obtain a video sequence of the bobsleigh start phase and, based on computer vision technology, identify and track the target to be analyzed in the video frames, thereby determining the trajectory information of the target to be analyzed. Video sequence analysis can be accurate to every frame and every action detail, thereby achieving in-depth analysis of the athlete's movements. Based on the detection of key points of the human body of the target to be analyzed, the key point information of the target to be analyzed is determined, and motion analysis of the detected target is performed based on the obtained data, realizing an automatic motion analysis process for the target to be analyzed, and achieving accurate identification and tracking of the athlete's movements. This improves the accuracy of the analysis results and reduces the subjectivity of human judgment. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the technical solutions in the present invention or the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0033] Figure 1 It is a flow chart of the motion analysis method for the sled departure phase provided by the present invention.

[0034] Figure 2 It is a schematic diagram of the analysis process provided by the present invention.

[0035] Figure 3 It is a schematic diagram of the time constraint flow provided by the present invention.

[0036] Figure 4 It is a schematic diagram of analysis result feedback provided by the present invention.

[0037] Figure 5 It is a schematic diagram of the setting of the image acquisition device provided by the present invention.

[0038] Figure 6 It is a schematic diagram of the target tracking process provided by the present invention.

[0039] Figure 7 It is a schematic diagram of the human body key point information extraction process provided by the present invention.

[0040] Figure 8 The present invention is a schematic diagram of the structure of a device for analyzing the motion of a sled during its departure phase.

[0041] Figure 9 It is a structural schematic diagram of the motion analysis device for the sled departure phase provided by the present invention.

[0042] Figure 10 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0043] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of the present invention.

[0044] This approach offers new possibilities for analyzing bobsleigh motion postures. Related methods primarily employ techniques such as human contour extraction, machine learning-based motion recognition, and depth-based 3D reconstruction. Accurately locating the resolution points due to obstructions is a common limitation of video analysis. By synthesizing the video panorama of the resolution points and utilizing adjacent frames within the same video to determine the resolution points, image analysis and automatic tracking based on the motion trajectory of key points can be performed more clearly, making 3D analysis more scientific. However, the challenge is that real-time data analysis during training and competition cannot be performed to provide coaches with timely feedback, complicating the application of the equipment. Machine learning in related methods struggles to accurately capture the complex nonlinear structure of the human body and the spatiotemporal relationships between consecutive motion frames. In practice, the human body is simplified into a rigid body connected by a series of rotating joints to identify and capture motion. However, human motion is more complex, and organs such as bones and muscles cannot be represented by simple rigid body models. Furthermore, machine learning algorithms require complex kinematic data preprocessing procedures and have limited processing power, making it difficult to accurately analyze bobsleigh technical movements in a timely manner.

[0045] To scientifically collect athlete parameters under practical conditions, motion capture technology is widely used in bobsleigh analysis. Related methods include infrared motion capture, inertial wearable motion capture, and video analysis. Infrared motion capture is expensive, has a complex operating process, and is affected by light, making it unsuitable for real-time bobsleigh monitoring. Inertial wearable motion capture uses wearable devices, which are generally difficult to wear and can interfere with bobsleigh athletes' training. They can also cause signal interference, no video footage, no displacement indicators, and low accuracy. Related methods include video analysis, which uses high-speed cameras to record videos and then processes them using software, requiring a significant amount of manual work.

[0046] Furthermore, the unique characteristics of bobsleigh, such as high instantaneous speeds, highly dispersed tracks, long race distances, and low-temperature, non-laboratory venues, present significant challenges in biomechanical analysis. To address this unique sport and environment, wind tunnel flow field simulations and computational simulations provide a theoretical basis for training. For example, wind tunnel simulations measure the biomechanical parameters of athletes' body posture during the start-up and acceleration phases, revealing that incoordination in the ice-breaking motion can affect gliding smoothness, thereby helping athletes correct incorrect techniques. Based on wind tunnel testing, Momose et al. conducted wind resistance tests for four bobsleigh postures: lying flat, head tilted, left / right head tilted, and foot lifted. They also conducted CFD simulations of the aerodynamic characteristics of the human / bobsleigh multibody system during the gliding phase to explore the theoretically optimal ice-breaking gliding posture. In bobsleigh events, when the head is raised to control and adjust course, intentional head tilting can be used to minimize the adverse effects of wind resistance. Bobsleigh competition is highly competitive, with races timed to 0.001 seconds. A 10% reduction in wind resistance during the gliding phase can improve performance by 1%. Pointing or curling the toes can result in a 27% difference in CDA, significantly impacting performance. Athletes should maintain a tight toeline during skating, but the theoretical research presented above is relatively idealistic. To further explore real-world sport scenarios, related methods have also applied video panoramic analysis technology and ultra-wideband (UWB)-based biomechanical systems to scientific bobsleigh measurements. However, video panoramic analysis requires a large amount of manual labeling and is slow to process, making it difficult to provide real-time feedback for daily training. UWB-based biomechanical systems require athletes to wear sensors throughout the entire motion, which still has some impact on the aerodynamic characteristics of the human / bobsleigh multi-body system at all stages.

[0047] In view of the shortcomings of the related methods, the present invention provides a motion analysis method for the sled start phase. Figure 1 FIG. 1 is a flow chart of the motion analysis method for the sled departure phase provided by the present invention. Figure 1 As shown, the method includes the following:

[0048] Step 110 , obtaining a video sequence of the target to be analyzed during the sledding start phase;

[0049] Step 120: identifying the target to be analyzed in each video frame in the video sequence, and determining motion trajectory information of the target to be analyzed based on the identified target to be analyzed in each video frame;

[0050] Step 130, performing human body key point detection on the target to be analyzed in each video frame to obtain human body key point information of the target to be analyzed in each video frame;

[0051] Step 140 : Based on the motion trajectory information, the key point information of the target to be analyzed in each video frame, and the time information of the target to be analyzed during the sled departure phase, a motion analysis is performed on the target to be analyzed to determine a motion analysis result of the target to be analyzed.

[0052] The executor of the bobsleigh start phase motion analysis method provided by the present invention can be an electronic device, a component of an electronic device, an integrated circuit, or a chip. The electronic device can be a mobile electronic device or a non-mobile electronic device. Exemplary mobile electronic devices include mobile phones, tablet computers, laptop computers, PDAs, ultra-mobile personal computers (UMPCs), netbooks, or personal digital assistants (PDAs), while non-mobile electronic devices include servers, network attached storage (NAS), or personal computers (PCs), etc., although the present invention does not impose specific limitations thereon.

[0053] The following describes the technical solution of the present invention in detail by taking the computer execution of the motion analysis method for the sled start phase provided by the present invention as an example.

[0054] In step 110 , a video sequence of the target to be analyzed during the sled departure phase of the sled movement is obtained.

[0055] Based on cameras at multiple angles, video data of the target during the sledding movement at the starting stage can be collected, and a video sequence can be constructed based on the collected video data.

[0056] Among them, the sled start phase during sledding can be divided into three parts, namely:

[0057] 1. Push and pull the handle to push the sled into the track;

[0058] 2. Use ice scraping gloves to paddle for about 7 meters;

[0059] 3. Change to the most aerodynamic posture.

[0060] In step 120, the target to be analyzed in each video frame in the video sequence is identified, and based on the identified target to be analyzed in each video frame, the motion trajectory information of the target to be analyzed is determined.

[0061] Computer vision techniques (such as deep learning models) are used to automatically identify athletes in videos. This typically involves object detection algorithms such as YOLO (You Only Look Once) and SSD (Single Shot MultiBox Detector), which can quickly and accurately locate the athlete's position within the video frame. By identifying the athlete in each frame, a continuous image sequence of the athlete throughout the starting phase can be constructed.

[0062] Once the athlete is identified in each frame, tracking algorithms can be used to calculate the athlete's trajectory. These algorithms analyze subtle changes in the athlete's position between frames to construct the athlete's movement path throughout the start phase. This trajectory information, including key parameters such as velocity, acceleration, and displacement, is crucial for understanding the athlete's movement patterns.

[0063] In step 130, human body key point detection is performed on the target to be analyzed in each video frame to obtain human body key point information of the target to be analyzed in each video frame.

[0064] Human keypoint detection is another important technology in computer vision. It accurately labels key parts of the human body (such as the head, shoulders, elbows, and knees) in video frames. This is typically achieved using deep learning models that can identify and locate multiple key points on the human body. During the bobsleigh launch phase, this information is crucial for assessing the athlete's posture, balance, and force distribution.

[0065] In step 140, based on the motion trajectory information, the body key point information of the target to be analyzed in each video frame, and the time information of the target to be analyzed during the sled departure phase, motion analysis is performed on the target to be analyzed to determine a motion analysis result of the target to be analyzed.

[0066] It's important to note that luge is a tactical sport in which athletes lie supine on a sled, controlling their center of gravity to correct their glide path on a specially designed ice track, and completing giant slalom turns in the shortest possible time. Starting speed is a key factor in luge performance, and starting time is the most effective indicator of an athlete's starting performance.

[0067] The specific analysis flow chart can be as follows Figure 2The analysis process provided by the present invention is shown in a schematic diagram. The motion trajectory information, key body point information, and time information (such as starting time and the time of key movements) collected in the above steps are integrated for comprehensive analysis. This can include evaluating key movements such as the athlete's compression angle at start, body lift angle, scraping amplitude, and scraping frequency. By comparing the data of standard movements or those of elite athletes, it is possible to further determine whether the target's performance meets the standard and where potential areas for improvement exist.

[0068] The bobsleigh start phase consists of three parts. Therefore, the speed of movement during pushing, pulling, and scraping, as well as the symmetry and stability of both limbs, are crucial factors in determining an athlete's performance. Focusing on the crucial phase of the bobsleigh start, the system, designed to meet the needs of bobsleigh athletes for competition preparation and competition, analyzes and visualizes data such as the athlete's start posture, speed, power, and gliding form in real time. This helps coaches and athletes instantly assess training status and movement accuracy, allowing them to correct deviations. It also summarizes each athlete's technical performance and training progress, providing a basis for coaches to develop personalized training plans.

[0069] For the moment information of the target to be analyzed during the sled start phase, a time constraint can be constructed. A time constraint is a precise time measurement of an event within a specific time period. In the sled start phase, the time constraint is used to record the time from the start signal to the athlete entering the track. The specific process can be as follows: Figure 3 The time constraint flow diagram provided by the present invention is shown as follows.

[0070] Specifically include:

[0071] High-precision timer:

[0072] Hardware: Use high-precision timers, such as photoelectric gates and high-speed timers, to record key time points. Trigger mechanism: Set up a trigger mechanism for the start signal to ensure that the timer starts timing when the signal is issued.

[0073] Video Sync:

[0074] Timestamp: Add timestamp to each camera's video frame to ensure that the video stream is consistent with the timer. Synchronization calibration: Ensure that the frame rate and timestamp of all cameras are synchronized through calibration to avoid time deviation.

[0075] Time Stamp:

[0076] Key time points: Mark key time points in the video, such as the start signal, the moment the athlete starts pushing the sled, the moment the athlete enters the track, etc. Time difference calculation: Calculate the difference between each time point and generate a time report.

[0077] Data Analysis:

[0078] Reaction time: The time from the start signal to the start of the sled push is calculated to evaluate the reaction speed. Pushing speed: The time from the start of the sled push to the start of the sled push is calculated to evaluate the pushing speed.

[0079] Comprehensive Report: Generate detailed performance reports including reaction time, sled pushing speed, key point analysis, etc. to help coaches and athletes optimize training and competition strategies

[0080] Based on the obtained motion analysis results, the training plan for the target to be analyzed can be supported. The specific process can be as follows Figure 4 The analysis result feedback diagram provided by the present invention is shown.

[0081] First, video images are captured using high-speed cameras. To minimize the impact of occlusion caused by the number and placement of cameras on capturing detailed image features, and to minimize computational limitations on motion capture accuracy, four high-speed cameras are used for fixed-range shooting. Second, the data captured by the hardware is fed into the software for analysis. A convolutional neural network deep learning model is primarily used to track key points of the human body in the images. Based on this key point data, key movement metrics such as compression angle, rise angle, ice scraping amplitude, and ice scraping frequency are calculated to support the training plan.

[0082] The motion analysis method for the bobsleigh start phase provided by the present invention obtains a video sequence of the bobsleigh start phase and, based on computer vision technology, identifies and tracks the target to be analyzed in the video frames, thereby determining the trajectory information of the target to be analyzed. Video sequence analysis can be accurate to every frame and every action detail, thereby achieving in-depth analysis of the athlete's movements. Based on the detection of key points of the human body of the target to be analyzed, the key point information of the target to be analyzed is determined, and motion analysis of the detected target is performed based on the obtained data, thus realizing an automatic motion analysis process for the target to be analyzed, and achieving accurate identification and tracking of the athlete's movements. This improves the accuracy of the analysis results and reduces the subjectivity of human judgment.

[0083] In one embodiment, obtaining a video sequence of a target to be analyzed during a sled departure phase during sled motion includes: synchronously obtaining video data of the target to be analyzed at multiple angles during the sled departure phase based on multiple image acquisition devices and moment information of the target to be analyzed during the sled departure phase; fusing the video data at the multiple angles to obtain fused video data, and extracting video frames from the fused video to obtain the video sequence.

[0084] Optionally, four image acquisition devices may be provided, and the location diagram may be as follows: Figure 5The image acquisition device provided by the present invention is shown in the schematic diagram. It is based on four image acquisition devices (cameras C1-C4) arranged around the starting stage of the sled, where G is the starting point of the target to be analyzed.

[0085] Based on the moment information of the target to be analyzed during the sled departure phase, an accurate timestamp is added to the video data recorded by each camera to ensure that video data from different angles can be accurately aligned to the same time point later.

[0086] Using computer vision techniques, we perform image registration on video frames from different cameras. This involves adjusting the position, rotation, and scale of the images so that they are aligned in the same spatial coordinate system. Based on the image registration results, we fuse and stitch the video data from multiple angles into a single video, generating fused video data.

[0087] From the fused video data, video frames are extracted according to a preset frame rate or key events (such as specific athlete movements). The extracted frames are arranged in chronological order to form a continuous video sequence. This video sequence is used to analyze the athletes' movements during the bobsleigh start phase in detail.

[0088] In a real-time example, based on the motion trajectory information, the human body key point information of the target to be analyzed in each video frame, and the moment information of the target to be analyzed during the sled departure phase, motion analysis is performed on the target to be analyzed to determine the motion analysis result of the target to be analyzed, including: determining the reaction speed of the target to be analyzed and the sled pushing speed of the target to be analyzed based on the motion trajectory information and the moment information; determining the motion angle data of the target to be analyzed based on the human body key point information of the target to be analyzed in each video frame, the motion angle data including at least joint angles, trunk compression angles, and body extension angles; and determining the motion analysis result of the target to be analyzed based on the reaction speed, the sled pushing speed, and the motion angle data.

[0089] Specifically, by analyzing the time between an athlete receiving the starting signal and starting to push the sled, the athlete's reaction speed can be calculated. This data is crucial for evaluating an athlete's agility and concentration.

[0090] Using the velocity data in the motion trajectory information, we can analyze the acceleration and final velocity of the athlete while pushing the sled. In particular, we focus on the speed changes of the athlete in the first few meters before departure.

[0091] Combining reaction speed, sled pushing speed, and movement angle data allows for a comprehensive assessment of the athlete's starting phase, including power output, movement efficiency, and body posture adjustments.

[0092] In one embodiment, identifying the target to be analyzed in each video frame in the video sequence includes: identifying the target to be analyzed in each video frame in the video sequence based on the target detection algorithm YOLO, and determining the target detection frame of the target to be analyzed in each video frame.

[0093] Determining the motion trajectory information of the target to be analyzed based on the target to be analyzed identified in each video frame includes: tracking the target detection frame in each video frame based on the target tracking algorithm Deep SORT to determine the trajectory information of the target to be analyzed.

[0094] Target tracking refers to the process of continuously detecting and locating a specific target in a video sequence. The target tracking process can be as follows: Figure 6 The target tracking process provided by the present invention is shown in the following schematic diagram. During the bobsleigh start phase, target tracking is used to monitor the entire process from the start to the entry of the runner onto the track. Specifically, it can be divided into the following three steps.

[0095] Object detection: Common detection algorithms include YOLO (You Only Look Once), SSD (Single Shot MultiBox Detector), and Faster R-CNN. YOLO, for example, is a real-time object detection algorithm that detects and locates objects with a single forward propagation. YOLO divides the input image into multiple grids, each of which is responsible for predicting a certain number of bounding boxes and their corresponding class probabilities.

[0096] Target recognition: Use convolutional neural networks (CNNs) to extract target features. Common networks include ResNet and VGG. The extracted features are input into a classifier to identify the target category.

[0097] Object tracking: Common tracking algorithms include Deep SORT, FairMOT, and Kalman Filter. DeepSORT, for example, combines deep learning and the Kalman filter to achieve robust multi-target tracking by maintaining a target's state estimate (position, velocity, etc.) and appearance features. In each frame, the detection result is matched with the target by calculating the similarity between the candidate detection and the existing target (such as the IoU and appearance feature distance), and the target's state is updated.

[0098] Based on the target tracking algorithm Deep SORT, the target detection box in each video frame is tracked to determine the trajectory information of the target to be analyzed.

[0099] In one embodiment, human key point detection is performed on the target to be analyzed in each video frame to obtain human key point information of the target to be analyzed in each video frame, including: based on HRNet key point detection technology, human key point detection is performed on the target to be analyzed in each video frame to obtain human key point information of the target to be analyzed in each video frame.

[0100] The extraction process of human body key point information can be as follows Figure 7 The human body key point information extraction process provided by the present invention is shown in the schematic diagram.

[0101] HRNet (High-Resolution Network) is a deep learning model designed for high-precision human keypoint detection. Traditional keypoint detection methods typically reduce the resolution of feature maps in the early stages to reduce computation and memory consumption, but this results in decreased keypoint detection accuracy. HRNet maintains high-resolution feature maps throughout the network and gradually introduces multi-scale feature fusion, improving the model's robustness and generalization capabilities while maintaining high accuracy.

[0102] The core idea of ​​HRNet is to maintain high-resolution feature maps throughout the network and enhance feature representation through multi-scale feature fusion. Specifically, HRNet consists of the following main parts:

[0103] Initial layer: The initial part of the network usually includes a convolutional layer and a maximum pooling layer to extract preliminary low-level features. The output feature map of this part has a low resolution but contains rich low-level information.

[0104] High-resolution branches: The core of HRNet is multiple high-resolution branches, which maintain high-resolution feature maps at different stages. Each high-resolution branch extracts features through multiple residual blocks, which can effectively learn complex feature representations.

[0105] Multi-scale feature fusion: To incorporate multi-scale information, HRNet adds new low-resolution branches at each stage and combines feature maps of different resolutions through cross-stage feature fusion. Specifically, the feature maps of the high-resolution branch are downsampled to the resolution of the low-resolution branch, while the feature maps of the low-resolution branch are upsampled to the resolution of the high-resolution branch. In this way, the high-resolution branch can utilize the contextual information of the low-resolution branch, while the low-resolution branch can benefit from the detailed information of the high-resolution branch.

[0106] Final layer: At the final stage of the network, all feature maps of different resolutions are fused into a high-resolution feature map. This high-resolution feature map passes through a series of convolutional layers and upsampling operations to ultimately generate a heat map, where each pixel value represents the probability of that location being a keypoint.

[0107] Based on the HRNet key point detection technology, human key point detection is performed on the target to be analyzed in each video frame to obtain the human key point information of the target to be analyzed in each video frame.

[0108] The present invention also provides a device using the motion analysis method for the sled departure phase provided by the present invention.

[0109] The device mainly consists of three parts: one is the video data acquisition end, which is mainly composed of high-definition acquisition equipment (four monocular cameras); the second is the three-dimensional data analysis software platform end; and the third is the display device end, which is mainly composed of touch-screen all-in-one machines in the coach's seat and the athlete's channel.

[0110] The specific implementation process can be as follows Figure 8 The schematic diagram of the device structure using the motion analysis method for the sled departure phase provided by the present invention is shown.

[0111] First, we used 3D video motion capture and video feedback technology to capture markerless video images of the bobsled athletes' technical movements during the starting phase. The acquisition process included two steps: video image acquisition, the construction of a human motion and posture recognition model, and the application of a machine learning algorithm to identify image features. Key parameter data was then calculated and evaluated. The details are as follows:

[0112] Video images were captured using high-speed cameras. The number and placement of cameras were designed to balance the effects of occlusion caused by perspective changes on image detail capture and computational limitations on motion capture accuracy, ensuring that they were tailored to the biomechanical characteristics and motion scenario of the target. Four high-speed cameras were used to capture the athletes' starting phase, the moment of departure, and the scrambling motion. The angles of the four principal optical axes ranged from 60° to 120°, ensuring that the entire range of the athletes' movements was captured. Once fixed, the cameras were not adjusted until filming was complete. The frame rate was set to 60 Hz, the shutter speed to 1 / 1000 s, and manual focus was used. Before filming, a universal calibration frame was set at the center of the athletes' range of motion to ensure that the frame was fully captured within the four cameras' frames, with an optimal occupancy ratio of 80% to 90%. The bobsleigh team athletes' technical movements during the starting phase were captured using a high-definition camera system. Video analysis was performed to analyze trunk compression and body lift angles, as well as upper limb explosive power. Kinematic characteristics of the bobsleigh athletes' starting and scrambling motions, including scrambling power, scrambling height, scrambling frequency, scrambling amplitude, and arm length, were also analyzed. A high-definition camera is used to perform three-dimensional plane shooting at the starting position. The camera angle is perpendicular to the athlete's movement direction. The upper torso midline and horizontal compression angle and the upper torso midline and horizontal upright angle of the athlete at the start are analyzed.

[0113] 3D video analysis extracts kinematic parameters from movements captured by four cameras. Using machine learning techniques, the collected athlete data is analyzed and presented visually, including charts, angle curves, and starting speeds. This provides statistical data on athletes' training status, performance, and workload across different cycles, summarizing their technical performance and training progress, providing a basis for coaches to develop personalized training plans. Finally, an intelligent storage and analysis system helps coaches build a large database of athletes, categorizing and storing training videos and data. Comparing the training performance of the same athlete over time allows for effective assessment of training effectiveness at each stage. Comparing the training performance of different athletes facilitates identifying gaps and implementing targeted improvements. This helps track athletes' performance throughout their developmental stages.

[0114] The advantages of this device are: to avoid invasive equipment interfering with the training status of athletes, biomechanical analysis can meet the requirements of transparent and non-interference collection. Sports cameras or high-speed cameras shoot the movement process in a fixed area, and perform motion recognition on the captured images and videos. Motion recognition based on images and videos can obtain more comprehensive kinematic information. In addition to the sports silhouettes of athletes at different times, it can also obtain biomechanical parameters such as joint angles, movement posture characteristics, and speed of athletes at different times. Traditional manual labeling and machine learning data processing have limited computing power. Capturing human motion features in captured images is a large task and time-consuming, which to a certain extent limits the analysis and rapid feedback of technical features of movements. The deep learning adopted by this device has representation advantages and flexibility compared to traditional methods. Among various deep learning neural network models, its multi-tasking processing capabilities can perceive the dynamic characteristics of the spatial domain.

[0115] The motion analysis device for the sled start phase provided by the present invention is described below. The motion analysis device for the sled start phase described below and the motion analysis method for the sled start phase described above can be referenced to each other.

[0116] like Figure 9 As shown, the device includes:

[0117] The video sequence acquisition module 910 is used to acquire a video sequence of the target to be analyzed during the sled departure phase of the sled movement;

[0118] a trajectory determination module 920 for identifying the target to be analyzed in each video frame in the video sequence, and determining motion trajectory information of the target to be analyzed based on the identified target to be analyzed in each video frame;

[0119] A key point determination module 930 is configured to perform human key point detection on the target to be analyzed in each video frame to obtain human key point information of the target to be analyzed in each video frame;

[0120] The motion analysis module 940 is configured to perform motion analysis on the target to be analyzed based on the motion trajectory information, the key point information of the target to be analyzed in each video frame, and the time information of the target to be analyzed during the sled departure phase, and determine a motion analysis result of the target to be analyzed.

[0121] The motion analysis device for the bobsleigh start phase provided by the present invention obtains a video sequence of the bobsleigh start phase and, based on computer vision technology, identifies and tracks the target to be analyzed in the video frames, thereby determining the trajectory information of the target to be analyzed. Video sequence analysis can be accurate to every frame and every action detail, thereby achieving an in-depth analysis of the athlete's movements. Based on the detection of key points of the human body of the target to be analyzed, the key point information of the target to be analyzed is determined, and motion analysis of the detected target is performed based on the obtained data, thereby realizing an automatic motion analysis process for the target to be analyzed, and achieving accurate identification and tracking of the athlete's movements. This improves the accuracy of the analysis results and reduces the subjectivity of human judgment.

[0122] In one embodiment, the video sequence acquisition module 910 is specifically configured to:

[0123] Obtain a video sequence of the target to be analyzed during the sled departure phase of the sled movement, including:

[0124] Synchronously acquiring video data of the target to be analyzed at multiple angles during the sled departure phase based on multiple image acquisition devices and time information of the target to be analyzed during the sled departure phase;

[0125] The video data from the multiple angles are fused to obtain fused video data, and video frames are extracted from the fused video to obtain the video sequence.

[0126] In one embodiment, the motion analysis module 940 is specifically configured to:

[0127] Performing motion analysis on the target to be analyzed based on the motion trajectory information, key point information of the target to be analyzed in each video frame, and time information of the target to be analyzed during the sled departure phase to determine a motion analysis result of the target to be analyzed, including:

[0128] Determining a reaction speed of the target to be analyzed and a sled pushing speed of the target to be analyzed based on the motion trajectory information and the time information;

[0129] Determining motion angle data of the target to be analyzed based on the human body key point information of the target to be analyzed in each video frame, wherein the motion angle data at least includes joint angles, trunk compression angles, and body extension angles;

[0130] A motion analysis result of the target to be analyzed is determined based on the reaction speed, the sled speed, and the motion angle data.

[0131] In one embodiment, the trajectory determination module 920 is specifically configured to:

[0132] Identifying the target to be analyzed in each video frame in the video sequence includes:

[0133] Based on the target detection algorithm YOLO, the target to be analyzed in each video frame in the video sequence is identified, and the target detection frame of the target to be analyzed in each video frame is determined.

[0134] In one embodiment, the trajectory determination module 920 is further configured to:

[0135] Determining motion trajectory information of the target to be analyzed based on the target to be analyzed identified in each video frame includes:

[0136] Based on the target tracking algorithm Deep SORT, target tracking is performed on the target detection frame in each video frame to determine the trajectory information of the target to be analyzed.

[0137] In one embodiment, the key point determination module 930 is specifically configured to:

[0138] Performing human key point detection on the target to be analyzed in each video frame to obtain human key point information of the target to be analyzed in each video frame includes:

[0139] Based on the HRNet key point detection technology, human key point detection is performed on the target to be analyzed in each video frame to obtain the human key point information of the target to be analyzed in each video frame.

[0140] Figure 10 An example of a physical structure diagram of an electronic device is shown below. Figure 10 As shown, the electronic device may include: a processor 1010, a communications interface 1020, a memory 1030, and a communication bus 1040, wherein the processor 1010, the communications interface 1020, and the memory 1030 communicate with each other via the communication bus 1040. The processor 1010 may call logic instructions in the memory 1030 to execute a motion analysis method for a sled start phase, the method comprising: acquiring a video sequence of a target to be analyzed during the sled start phase during sled motion;

[0141] Identifying a target to be analyzed in each video frame in the video sequence, and determining motion trajectory information of the target to be analyzed based on the identified target to be analyzed in each video frame;

[0142] Performing human body key point detection on the target to be analyzed in each video frame to obtain human body key point information of the target to be analyzed in each video frame;

[0143] Based on the motion trajectory information, the human body key point information of the target to be analyzed in each video frame, and the time information of the target to be analyzed during the sled departure phase, motion analysis is performed on the target to be analyzed to determine a motion analysis result of the target to be analyzed.

[0144] Furthermore, the logic instructions in the aforementioned memory 1030 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, a mobile hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0145] On the other hand, the present invention further provides a computer program product, the computer program product including a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can perform the motion analysis method for the sled start phase provided by the above methods, the method including: obtaining a video sequence of the sled start phase of a target to be analyzed during sled motion;

[0146] Identifying a target to be analyzed in each video frame in the video sequence, and determining motion trajectory information of the target to be analyzed based on the identified target to be analyzed in each video frame;

[0147] Performing human body key point detection on the target to be analyzed in each video frame to obtain human body key point information of the target to be analyzed in each video frame;

[0148] Based on the motion trajectory information, the human body key point information of the target to be analyzed in each video frame, and the time information of the target to be analyzed during the sled departure phase, motion analysis is performed on the target to be analyzed to determine a motion analysis result of the target to be analyzed.

[0149] In another aspect, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the computer program is implemented to perform the motion analysis method of the sled start phase provided by the above methods, the method comprising: obtaining a video sequence of the sled start phase of a target to be analyzed during sled motion;

[0150] Identifying a target to be analyzed in each video frame in the video sequence, and determining motion trajectory information of the target to be analyzed based on the identified target to be analyzed in each video frame;

[0151] Performing human body key point detection on the target to be analyzed in each video frame to obtain human body key point information of the target to be analyzed in each video frame;

[0152] Based on the motion trajectory information, the human body key point information of the target to be analyzed in each video frame, and the time information of the target to be analyzed during the sled departure phase, motion analysis is performed on the target to be analyzed to determine a motion analysis result of the target to be analyzed.

[0153] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one location or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0154] Through the description of the above embodiments, those skilled in the art will clearly understand that each embodiment can be implemented using software plus a necessary general-purpose hardware platform, or of course, hardware. Based on this understanding, the essence of the above technical solution, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, or an optical disk, and includes a number of instructions for causing a computer device (such as a personal computer, server, or network device) to execute the methods described in each embodiment or certain portions of the embodiments.

[0155] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A motion analysis method for a sled during its start phase, characterized in that: include: Obtain a video sequence of the target to be analyzed during the sled departure phase of the sled movement; Identifying a target to be analyzed in each video frame in the video sequence, and determining motion trajectory information of the target to be analyzed based on the identified target to be analyzed in each video frame; Performing human body key point detection on the target to be analyzed in each video frame to obtain human body key point information of the target to be analyzed in each video frame; Based on the motion trajectory information, the human body key point information of the target to be analyzed in each video frame, and the time information of the target to be analyzed during the sled departure phase, motion analysis is performed on the target to be analyzed to determine a motion analysis result of the target to be analyzed. Obtain a video sequence of the target to be analyzed during the sled departure phase of the sled movement, including: Synchronously acquiring video data of the target to be analyzed at multiple angles during the sled departure phase based on multiple image acquisition devices and time information of the target to be analyzed during the sled departure phase; The video data from the multiple angles are fused to obtain fused video data, and video frames are extracted from the fused video to obtain the video sequence.

2. The motion analysis method for the sled start phase according to claim 1, characterized in that: The step of performing motion analysis on the target to be analyzed based on the motion trajectory information, the key point information of the target to be analyzed in each video frame, and the time information of the target to be analyzed during the sled departure phase, and determining a motion analysis result of the target to be analyzed includes: Determining a reaction speed of the target to be analyzed and a sled pushing speed of the target to be analyzed based on the motion trajectory information and the time information; Determining motion angle data of the target to be analyzed based on the human body key point information of the target to be analyzed in each video frame, wherein the motion angle data at least includes joint angles, trunk compression angles, and body extension angles; A motion analysis result of the target to be analyzed is determined based on the reaction speed, the sled speed, and the motion angle data.

3. The motion analysis method for the sled start phase according to claim 1, characterized in that: The identifying the target to be analyzed in each video frame in the video sequence includes: Based on the target detection algorithm YOLO, the target to be analyzed in each video frame in the video sequence is identified, and the target detection frame of the target to be analyzed in each video frame is determined.

4. The motion analysis method for the sled start phase according to claim 3, characterized in that: The determining of motion trajectory information of the target to be analyzed based on the target to be analyzed identified in each video frame includes: Based on the target tracking algorithm Deep SORT, target tracking is performed on the target detection frame in each video frame to determine the trajectory information of the target to be analyzed.

5. The motion analysis method for the sled start phase according to claim 1, characterized in that: The performing of human key point detection on the target to be analyzed in each video frame to obtain human key point information of the target to be analyzed in each video frame includes: Based on the HRNet key point detection technology, human key point detection is performed on the target to be analyzed in each video frame to obtain the human key point information of the target to be analyzed in each video frame.

6. A motion analysis device for a sled during its departure phase, characterized in that: include: A video sequence acquisition module is used to acquire a video sequence of the target to be analyzed during the sled departure phase of the sled movement; a trajectory determination module, configured to identify the target to be analyzed in each video frame in the video sequence, and determine motion trajectory information of the target to be analyzed based on the identified target to be analyzed in each video frame; A key point determination module is used to detect the human key points of the target to be analyzed in each video frame to obtain the human key point information of the target to be analyzed in each video frame; a motion analysis module, configured to perform motion analysis on the target to be analyzed based on the motion trajectory information, the key point information of the target to be analyzed in each video frame, and the time information of the target to be analyzed at the sled departure stage, and determine a motion analysis result of the target to be analyzed; Obtain a video sequence of the target to be analyzed during the sled departure phase of the sled movement, including: Synchronously acquiring video data of the target to be analyzed at multiple angles during the sled departure phase based on multiple image acquisition devices and time information of the target to be analyzed during the sled departure phase; The video data from the multiple angles are fused to obtain fused video data, and video frames are extracted from the fused video to obtain the video sequence.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the motion analysis method for the sled start phase according to any one of claims 1 to 5 is implemented.

8. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the motion analysis method for the sled start phase according to any one of claims 1 to 5 is implemented.

9. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the motion analysis method for the sled start phase according to any one of claims 1 to 5 is implemented.

Images