MeMoMan - model based markerless capturing of human motion
The model-based markerless motion capturing system MeMoMan is currently being developed at the Institute of Ergonomics and the Institute for Image Understanding & Knowledge-Based Systems of the Technische Universit.t München. The basic concept is to combine a detailed digital human model with state of the art image processing algorithms. Based on standard video data motion capturing is performed in a Bayesian framework using a set of hierarchically coupled local particle filters. The algorithms have been extended with knowledge on motion control to improve accuracy and reduce computation time. Capturing different tasks under laboratory conditions showed the use and robustness of the approach. Future work will evaluate the motion capturing quality based on ground truth data from a marker-based motion capturing system. Improvement of the algorithms and integration of motion models will further extend the motion capturing to a wider range of applications and make it more independent from environmental conditions.
Florian Engstler Jan Bandouch Heiner Bubb
Institute of Ergonomics, Technische Universit(a)t München, Boltzmannstr. 15, 85747 Garching Institute for Image Understanding & Knowledge-Based Systems, Technische Universit(a)t München,Boltzm
国际会议
17th World Congress on Ergonomics(第十七届国际人类工效学大会)
北京
英文
1-6
2009-08-09（万方平台首次上网日期，不代表论文的发表时间）