Skill Movement Trajectory Recognition of Freestyle Skiing U-Shaped Field Based on Deep Learning and Multitarget Tracking Algorithm

Abstract

Freestyle skiing U-shaped field is a snow sport that uses double boards to perform a series of action skills in a U-shaped pool, which requires very high skills for athletes. In this era of deep learning, in order to develop a more scientific training method, this paper combines multitarget tracking algorithm and deep learning to conduct research in freestyle skiing U-shaped venue skills motion capture. Therefore, this paper combines the convolutional neural network and multitarget tracking algorithm in deep learning to study the human action recognition technology, and then uses the LSTM module to study the freestyle skiing U-shaped venue skills. Finally, this paper designs the training method of the action recognition algorithm of the freestyle U-shaped skiing skills multitarget tracking algorithm based on deep learning. This paper also designs multitarget tracking dataset experiments and model updating experiments. Based on the data of experimental analysis, the training method designed in this paper is optimized, and finally compared with the traditional training method. Compared with the traditional freestyle U-shaped skiing skills training method, the experimental results show that the training method of the freestyle U-shaped skiing skills multitarget tracking algorithm action recognition algorithm is based on deep learning designed in this paper and this improves the skill score by 14.48%. Most professional students are very satisfied with the training method designed in this paper.

Motion Tracking of Daily Living and Physical Activities in Health Care: Systematic Review From Designers’ Perspective

“…The role of AI mainly functions for data classification, monitoring, and visualization among the included studies. Furthermore, 80% (43/54) of the studies recruited participants for user study [ 75 , 76 , 78 - 81 , 83 - 88 , 90 , 92 , 93 , 95 - 105 , 107 - 110 , 113 - 116 , 118 - 124 , 127 , 128 ]. Twelve articles used public data set [ 78 , 82 , 89 , 91 , 94 , 106 , 111 , 112 , 117 , 125 - 127 ].…”

“…Furthermore, 80% (43/54) of the studies recruited participants for user study [ 75 , 76 , 78 - 81 , 83 - 88 , 90 , 92 , 93 , 95 - 105 , 107 - 110 , 113 - 116 , 118 - 124 , 127 , 128 ]. Twelve articles used public data set [ 78 , 82 , 89 , 91 , 94 , 106 , 111 , 112 , 117 , 125 - 127 ]. Most of the public data set that were used are University of California, Irvine human activity recognition, center of advanced studies in adaptive system, mobile health, University of Milano Bicocca smartphone-based human activity recognition, mobile sensor data anonymization, a large multipurpose human motion and video data set, and wireless sensor data mining data sets.…”

“…They were used to collect data on typical ADLs, such as standing, walking, sitting, and jumping, while ambient sensors were used less frequently. Physical activities such as jumping [ 75 , 78 , 98 ], playing basketball [ 96 , 116 ], soccer [ 83 ], baseball [ 115 ], running [ 105 ], ballet dance [ 103 ] skiing [ 127 ], karate [ 84 ], and Taijiquan [ 123 ] were studied. Meanwhile, AI processing in the included articles functions for data extraction and classification to achieve daily living and physical activities recognition.…”

“…The investigations involving the placement of sensors in physical spaces mostly use motion capture and image processing techniques [ 117 , 119 ]. Researchers have placed high-speed cameras to capture movements during sports activities [ 75 , 82 - 84 , 88 , 98 , 101 , 105 , 115 , 120 , 127 ] and in specialized work situations, such as maritime occupations [ 104 ]. One study [ 93 ] placed accelerometer and gyroscope sensors on an object, specifically a water container, to determine the frequency and amount of use from the container’s motion tracking data.…”

“…Moreover, 13% (7/54) of the studies collected data from 50 to 100 individuals. Half (27/54, 50%) of the studies used 10 to 100 samples from public databases [ 77 , 89 , 94 , 106 , 111 , 112 , 117 , 125 - 127 ]. Only one study [ 107 ] amassed gait data from >1000 healthy individuals.…”