A Novel Marker Tracking Method Based on Extended Kalman Filter for Multi-Camera Optical Tracking Systems

Li Liu,Bo Sun,Ning Wei,Chao Hu
DOI: https://doi.org/10.1109/icbbe.2011.5780295
2011-01-01
Abstract:In the Robotics Assisted Surgical System, the position and orientation of the surgical tools must be estimated precisely in real time. In order to decrease the possibility of the occlusion of line-of-sight, the multi-camera optical tracking system prototype is developed. Based on this hardware platform, a novel multi-marker tracking algorithm using Extended Kalman Filter (EKF) for multi-camera tracking systems is proposed, which makes full use of the redundant information of multi-camera. The marker target movement model can benefit from multiple views by performing a special EKF updating 3D state by 2D projection observations on the image planes. This method has the advantage to estimate the movement of a 3D point more accurately by directly using the actual observations from multiple views rather than the non-continuous and possibly error-prone 3D reconstructed point. The experimental results indicate that the presented tracking algorithm is able to track the 3D trajectories of multiple targets simultaneously, and the estimated 3D positions by EKF are in agreement with the actual measurements by stereo triangulation.
What problem does this paper attempt to address?