@inproceedings{2189c46d45e54b58842c55e5cfbcd5bc,
title = "Human gait capture by tracking joint motion based on Kalman Filter",
abstract = "Human gait capture by vision plays an important role in the guidance for the rehabilitation of the lower limbs dysfunction patients. Unfortunately, it is not included in most of the present rehabilitation devices. This paper presents a novel real-time system for human gait capture by detecting the joints trajectory to obtain 3-D information of the joints motion by Kinect. Firstly, we put markers on the joints of each leg and detect them in a way combined background subtraction and multi-object template matching. Secondly, we estimate the potential regions in next frame by using Kalman Filter with current knowledge of the detected markers captured in the Kinect frame. Thirdly, we determine the position of markers within the potential region within the depth information. Finally, we accomplish gait capture by iteratively applying the estimation and detection method in a consecutive sequence of frames. This system achieves good tracking result and gait information at a frame rate of 30fps with a resolution of 640x480.",
keywords = "Gait Capture, Human Joints Detection, Kalman Filter, Kinect",
author = "Shuai Tian and Weihai Chen and Xingming Wu",
note = "Publisher Copyright: {\textcopyright} 2014 IEEE.; 9th IEEE Conference on Industrial Electronics and Applications, ICIEA 2014 ; Conference date: 09-06-2014 Through 11-06-2014",
year = "2014",
month = oct,
day = "20",
doi = "10.1109/ICIEA.2014.6931406",
language = "英语",
series = "Proceedings of the 2014 9th IEEE Conference on Industrial Electronics and Applications, ICIEA 2014",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "1499--1504",
booktitle = "Proceedings of the 2014 9th IEEE Conference on Industrial Electronics and Applications, ICIEA 2014",
address = "美国",
}