In this paper the problem of active object recognition/pose estimation is investigated. The principle component analysis is used to produce an observation vector from images captured simultaneously by multiple cameras from different view angles of an object belonging to a set of a priori known objects. Models of occlusion and sensor noise have been incorporated into a probabilistic model of sensor/object to increase the robustness of the recognition process with respect to such uncertainties. A recursive Bayesian state estimation problem is formulated to identify the object and estimate its pose by fusing the information obtained from the cameras at multiple steps. In order to enhance the quality of the estimates and to reduce the number of images taken, the positions of the cameras are controlled based on a statistical performance criterion, the Cramer-Rao lower bound (CRLB). Comparative Monte Carlo experiments conducted with a two-camera system demonstrate that the features of the proposed method, i.e. information fusion from multiple sources, active optimal sensor planing, and occlusion modelling are all highly effective for object classification/pose estimation in the presence of structured noise

This paper is concerned with the problem of appearance-based active multi-sensor object recognition/pose estimation in the presence of structured noise. It is assumed that multiple cameras acquire images from an object belonging to a set of known objects. An algorithm is proposed for optimal sequential positioning of the cameras in order to estimate the class and pose of the object from sensory observations. The principle component analysis is used to produce the observation vector from the acquired images. Object occlusion and sensor noise have been explicitly incorporated into the recognition process using a probabilistic approach. A recursive Bayesian state estimation problem is formulated that employs the mutual information in order to determine the best next camera positions based on the available information. Experiments with a two-camera system demonstrate that the proposed method is highly effective in object recognition/pose estimation in the presence of occlusion.