Title :
Whole-body pose estimation in physical rider-bicycle interactions with a monocular camera and a set of wearable gyroscopes
Author :
Xiang Lu ; Kaiyan Yu ; Yizhai Zhang ; Jingang Yi ; Jingtai Liu
Author_Institution :
Inst. of Robot. & Autom. Inf. Syst., Nankai Univ., Tianjin, China
Abstract :
We report the development of a human whole-body pose estimation scheme with application to rider-bicycle interactions. The estimation scheme is built on the fusion of measurements of a monocular camera on the bicycle and a set of small wearable gyroscopes attached to the rider´s upper- and lower-limb and the trunk. A single feature point is collocated with each wearable gyroscope and also on the segment link where the gyroscope is not attached. An extended Kalman filter is designed to fuse the vision-inertial measurements to obtain accurate whole-body poses. The estimation design also incorporates a set of constraints from human anatomy and the physical rider-bicycle interactions. We demonstrate and compare the performance of the estimation design through multiple subjects riding experiments.
Keywords :
Kalman filters; cameras; computer vision; gyroscopes; image fusion; nonlinear filters; pose estimation; EKF-based vision-inertial fusion scheme; extended Kalman filter; human whole-body pose estimation; monocular camera; physical rider-bicycle interactions; vision-inertial measurements; wearable gyroscopes; Bicycles; Cameras; Estimation; Gyroscopes; Joints; Quaternions; Wrist;
Conference_Titel :
Intelligent Robots and Systems (IROS 2014), 2014 IEEE/RSJ International Conference on
Conference_Location :
Chicago, IL
DOI :
10.1109/IROS.2014.6943143