In the following we assume that the random vector wk captures uncertainties in the model and vk denotes the measurement noise. Both are temporally uncorrelated (white noise), zero-mean random sequences with known covariances and both of them are uncorrelated with the initial state x0. E[wk] = 0 E[wkw T k ] = Qk E[wkw T j ] = 0 for k 6= j E[wkx T 0 ] = 0 for all k (3) E[vk] = 0 E[vkv T k ] = Rk ...

Estimating the structure of the human face is a long studied and difficult task. In this paper we present a new method for estimating facial structure from only a minimal number of salient feature points across a video sequence. The presented method uses both an Extended Kalman Filter (EKF) and a Kalman Filter (KF) to regress 3D Morphable Model (3DMM) shape parameters and solve 3D pose using a ...

We describe a model-based object tracking system that updates the connguration parameters of an object model based upon information gathered from a sequence of monocular images. Realistic object and imaging models are used to determine the expected visibility of object features, and to determine the expected appearance of all visible features. We formulate the tracking problem as one of paramet...

Abstract: This paper describes an automatic calibration procedure adopted to improve the localization of an outdoor mobile robot. The proposed algorithm estimates, by using an extended Kalman filter, the main kinematic parameters of the vehicles, such as the wheel radii and the wheelbase as well as the magnetometer offset. Several trials have been performed to validate the proposed strategy on ...

Abstract. A three-mode nonlinear slow-fast system with the fast forcing is studied here as a model for filtering turbulent signals from partial observations. The model describes the interaction of two externally driven fast modes with a slow mode through catalytic nonlinear coupling. The special structure of the nonlinear interaction allows for the analytical solution for the first and second o...

The Extended Kalman Filter (EKF) can be used for parameter as well as for state estimation. However, when used for parameter estimation, EKF is sensitive to modeling uncertainties and prone to instability. In this paper, the Variable Structure filtering concept is introduced and used for modifying the EKF into a robust form. The proposed method is model-based and has a guaranteed stability give...

The posterior Cramér-Rao lower bound (PCRB) for estimating the position and velocity of a Mobile Station (MS) by integrating two different kinds of measurement from GSM networks, Timing Advance (TA) and Received Signal Strength (RSS), is derived. The bound shows that theoretically the data fusion approach yields better accuracy than using a single kind of measurement. The performance of an exte...

In this article we propose a method for estimating the camera motion from a video-sequence acquired in the presence of general 3D structures. Solutions to this problem are commonly based on the tracking of point-like features, as they usually back-project onto viewpoint-invariant 3D features. In order to improve the robustness, the accuracy and the generality of the approach, we are interested ...

This paper presents a practical technique for modelbased 3D hand tracking. An anatomically accurate hand model is built from truncated quadrics. This allows for the generation of 2D profiles of the model using elegant tools from projective geometry, and for an efficient method to handle self-occlusion. The pose of the hand model is estimated with an Unscented Kalman filter (UKF), which minimize...