Orient-2: a realtime wireless posture tracking system using local orientation estimation
Proceedings of the 4th workshop on Embedded networked sensors
Use of Body Model Constraints to Improve Accuracy of Inertial Motion Capture
BSN '10 Proceedings of the 2010 International Conference on Body Sensor Networks
Activity recognition using biomechanical model based pose estimation
EuroSSC'10 Proceedings of the 5th European conference on Smart sensing and context
Using egocentric vision to achieve robust inertial body tracking under magnetic disturbances
ISMAR '11 Proceedings of the 2011 10th IEEE International Symposium on Mixed and Augmented Reality
IEEE Transactions on Robotics
Hi-index | 0.00 |
Inertial tracking is still an area of active research, especially in the context of real-time human motion capture. Existing systems either propose loosely coupled tracking approaches, taking the resulting drawbacks into account, or they propose tightly coupled solutions limited to a fixed chain with few segments, but without any flexibility in changing the skeleton structure. Therefore, this paper proposes a generic approach for tracking arbitrary kinematic chains in a tightly coupled recursive filtering framework. The generic property as well as the tracking stability of the proposed system are demonstrated and initial experimental results concerning its accuracy are also presented.