Spatial-temporal consistent labeling of tracked pedestrians across non-overlapping camera views

Tracking people across multiple cameras with non-overlapping views is a challenging task, since their observations are separated in time and space and their appearances may vary significantly. This paper proposes a Bayesian model to solve the consistent labeling problem across multiple non-overlapping camera views. Significantly different from related approaches, ourmodel assumes neither people arewell segmented nor their trajectories across camera views are estimated. We formulate a spatial–temporal probabilisticmodel in the hypothesis space that consists the potentiallymatchedobjects between the exit field of view (FOV) of one camera and the entry FOV of another camera. A competitive major color spectrum histogram representation (CMCSHR) for appearance matching between two objects is also proposed. The proposed spatial–temporal and appearancemodels are unified by amaximum-a-posteriori (MAP) Bayesianmodel. Based on this Bayesianmodel, when a detected newobject corresponds to a group hypothesis (more than one object), we further develop an online method for online correspondence update using optimal graph matching (OGM) algorithm. Experimental results on three different real scenarios validate the proposedBayesianmodel approach and the CMCSHRmethod. The results also show that the proposed approach is able to address the occlusion problem/group problem, i.e. finding the corresponding individuals in another camera view for a group of people whowalk together into the entry FOV of a camera. & 2010 Elsevier Ltd. All rights reserved.