Video-based diving platform skiing motion analysis method
A technology for skiing and analysis methods, applied in the fields of instruments, character and pattern recognition, computer parts, etc., can solve the problems of less motion analysis, difficult implementation, and restrictions on ski jumping events, and achieves improved accuracy and simple data acquisition. Effect
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment Construction
[0053] Such as image 3 As shown, a video-based ski jumping motion analysis method, the method includes the following sequential steps:
[0054] (1) Acquire the video data of ski jumping: use the calibrated high-speed camera to collect the action data of the skiers, preprocess the data and make a data set, and use the labeling tool Labelme to identify the visible key nodes of the skiers and snowboards Label with the real bounding box Boundingbox information;
[0055] (2) construct athlete and snowboard detection network model, key node detection network model respectively: described athlete and snowboard detection network model adopt YOLOv5s deep convolutional neural network, described key node detection network model adopts HRNet deep convolutional neural network;
[0056] (3) train athlete and snowboard detection network model, key node detection network model respectively;
[0057] (4) Predict the key nodes of the athletes and snowboards in the video image: First, use the...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


