Title :
A Bayesian Network for online evaluation of sparse features based multitarget tracking
Author :
Biresaw, Tewodros ; Regazzoni, C.S.
Author_Institution :
Univ. of Genova, Opera, Italy
fDate :
Sept. 30 2012-Oct. 3 2012
Abstract :
Online evaluation of tracking algorithms has received attentions in computer vision community to detect failures and apply correction methods for achieving better performances. In this paper, a novel online evaluation framework is proposed for a multitarget feature points based object tracking. An online partial least square regression and correlation model is constructed from short trajectory histories for the tracks. The model allows to estimate the state of one track from the other track states. The core idea for the method is creating a virtual reference data for evaluation from the learned model. The proposed self-evaluation mechanism is presented as a Dynamic Bayesian Network. The method is evaluated on a simulation data for tracking feature points from a pedestrian.
Keywords :
belief networks; computer vision; correlation methods; feature extraction; least squares approximations; object detection; object tracking; regression analysis; target tracking; computer vision community; correlation model; dynamic Bayesian network; failure detection; multitarget feature point tracking; object tracking; online evaluation; online partial least square regression; pedestrian; self-evaluation mechanism; sparse features based multitarget tracking; tracking algorithm; trajectory histories; virtual reference data; Bayesian methods; Correlation; Heuristic algorithms; Indexes; Predictive models; Tracking; Trajectory; Dynamic Bayesian Network; Partial Least Square regression; Trajectory analysis; Visual tracking;
Conference_Titel :
Image Processing (ICIP), 2012 19th IEEE International Conference on
Conference_Location :
Orlando, FL
Print_ISBN :
978-1-4673-2534-9
Electronic_ISBN :
1522-4880
DOI :
10.1109/ICIP.2012.6466888
