This page presents a real-time, robust and efficient 3D model-based tracking algorithm adapted to catadioptric images. A virtual visual servoing approach is used for 3D tracking. This method is similar to more classical non-linear pose computation techniques. Robustness is obtained by integrating a M-estimator into the virtual visual control law via an iteratively re-weighted least squares implementation.
Results of the virtual visual servoing tracker with perspective cameras are presented in Real-time 3D localisation and tracking page.
In this page we present results of the extension of this approach to the use of catadioptric cameras.
click for the video
