Multiview image registration using augmented Kalman filter

Multiview image registration is widely used in computer vision applications. In this paper, an approach for multiview 3D image registration based on augmented Kalman filter is proposed by taking account of various uncertainties. The position and orientation of viewpoint are considered as system state. System augmentation model and system observation model are constructed. The position and orientation of each viewpoint is augmented and updated recursively. The global transformation parameters of image are computed with the state estimation of corresponding viewpoint. The proposed multiview image registration method can handle the uncertainty efficiently and the registration result is accurate and globally consistent. Some experimental results are provided to validate the performance of the proposed method.