3D Human Motion Simulation and a Video Analysis System for Sports Training

2005 ◽  
Vol 42 (2) ◽  
pp. 344 ◽  
Author(s):  
Zhaoqi Wang
2014 ◽  
Vol 926-930 ◽  
pp. 2743-2746 ◽  
Author(s):  
Rui Min Hu ◽  
Zhen Dong He ◽  
Feng Bai

With the rapid development of computer technology, human motion tracking based on video is a kind of using ordinary camera tracking unmarked human movement technology. It has important application value in automatic monitoring, human-computer interaction, sports analysis and many other fields. This research is a hot research direction in the field of computer vision in recent years. Because of the complexity of the problem and the lack of understanding of the nature of the human visual tracking based on video is always a difficult problem in computer vision. The research content of this article is set in sports training, for motion analysis of non-contact, no interfere with measurement and simulation requirements, the use of computer graphics and computer vision technology, discussing 3D human motion simulation technology based on video analysis.


2019 ◽  
pp. 425-440
Author(s):  
Rajan Bhatt ◽  
Kimberly Farrell ◽  
Karim Abdel-Malek ◽  
Jasbir Arora ◽  
Chris Murphy

Author(s):  
Shanshan Pu ◽  

The sports video analysis system in sports training can improve the ability of motion analysis and improve the training quality of sports training video playback. In view of the shortcomings of the current motion video analysis system, a new sports training video analysis system is proposed. The overall design of the video analysis system of the sports training system is analyzed, and the detailed design of the system is analyzed. Finally, the performance of the system is tested. The results show that the system can accurately analyze the video and image information of sports training. The accuracy of the key frame extraction is high and the recall rate is high. It can be used to guide the training of sports training.


Author(s):  
Woojin Park ◽  
Don B. Chaffin ◽  
Bernard J. Martin

Simulating human motions in the virtual CAD world is important in the computerized ergonomic design of products and workplaces. The present study introduces a novel, memory-based approach for simulating realistic human motions and presents a motion modification algorithm. in this novel approach, realistic human motions are simulated by modifying existing motion samples stored in a motion database. The proposed motion modification algorithm was found to be able to simulate human motions accurately. The memory-based motion simulation approach has advantages over existing simulation models as it can simulate qualitatively different types of motions on a single platform, predict motions of different styles, and continually learn new motions.


Author(s):  
Matthew P. Reed ◽  
Don B. Chaffin ◽  
Julian Faraway

Digital human figure models (DHM) are increasingly the tools of choice for assessments of the physical ergonomics of products and workplaces. Software representations of users and workers are used to visualize people performing tasks of interest. Analyses have usually focused on clearance and reach in static postures, not because the actual tasks are static, but rather because DHM have lacked robust, accurate motion simulation capability. Research is underway at many institutions to develop improved motion simulation methods, drawing on a wide variety of methodologies from fields such as computer graphics, kinesiology, motor control, and robotics. Experience in the Human Motion Simulation Laboratory at the University of Michigan suggests that conventional metrics of accuracy for posture and movement prediction do not adequately capture the aspects of human movement that are most important for ergonomic analysis. This paper identifies and justifies a set of these critical features.


Sign in / Sign up

Export Citation Format

Share Document