Abstract— GP-BayesFilters are a general framework for integrating Gaussian process prediction and observation models into Bayesian filtering techniques, including particle filt...
A requirement common to most dynamic vision applications is the ability to track objects in a sequence of frames. This problem has been extensively studied in the past few years, ...
Octavia I. Camps, Hwasup Lim, Cecilia Mazzaro, Mar...
— The unscented Kalman filter (UKF) has recently been proposed for filtering noisy chaotic signals. Though computationally advantageous, the UKF has not been thoroughly analyze...
We propose an HMM model for contour detection based on multiple visual cues in spatial domain and improve it by joint probabilistic matching to reduce background clutter. It is fu...