Skip to Main Content
In this paper we present a query by example approach for tracking people across non overlapping cameras. The method proposed is based on the use of a multi-dimensional feature vector and its covariance, defining an appearance model of every detected moving region in the network of cameras. The model uses relative pixel position, color and gradients descriptors of each detected object. Association of objects across non-overlapping cameras is performed by matching appearance of selected object with past observations. Availability of tracking within every camera can further improve the accuracy of such association by matching several targets appearance models with detected regions. For this purpose we present an automatic clustering technique allowing to build a multi-valued appearance model from a collection of covariance matrices. The proposed approach does not require geometric or colorimetric calibration of the cameras. We will illustrate the method for tracking people and objects in relatively crowded indoor scenes.