Quaternion-Based Tracking of Multiple Objects in Synchronized Videos

This paper presents a method for tracking multiple objects using multiple cameras that integrates spatial position, shape and color information to track object blobs. Given three known points on the ground, camera calibration is computed by solving a set of quaternionbased nonlinear functions rather than solving approximated linear functions. By using a quaternion-based method, we can avoid the singularity problem. Our method focuses on establishing correspondence between objects and templates as the objects come into view. We fuse the data from individual cameras using an Extended Kalman Filter (EKF) to resolve object occlusion. Results based on calibration via Tsai’s method as well as our method are presented. Our results show that integrating simple features makes the tracking effective, and that EKF improves the tracking accuracy when long term or temporary occlusion occurs.