Recursive Camera Resectioning with Unscented Particle Filter in Image Sequen Application to Video-based Augmented R

In this paper, we propose a new recursive framework for camera resectioning and apply it to off-line video-based augmented reality. Our algorithm is based on an unscented particle filter, which deals with non-linear dynamic systems without local linearization, and leads to more accurate results than other non-linear filters. The proposed approach has some desirable properties. It does not rely on closed-form solutions. It is fairly accurate and is easy to implement as compared with other nonlinear approaches. As a result, the proposed algorithm outperforms the standard camera resectioning algorithm. We verify this through experimentation using real image sequences.