Estimating the positions of a set of movingobjects captured from a network of cameras is still anopen problem in Computer Vision. In this paper, a distributedand real-time approach for tracking multipleobjects on multiple cameras is presented. A quantitativecomparison with six state-of-the-art methods hasbeen carried out on the publicly available PETS 2009data set, demonstrating the eectiveness of the algorithm.Moreover, the proposed method has been testedalso on a multi-camera soccer data set, showing its datafusion capabilities.
A distributed approach for real-time multi-camera multiple object tracking
BLOISI, Domenico Daniele;
2017-01-01
Abstract
Estimating the positions of a set of movingobjects captured from a network of cameras is still anopen problem in Computer Vision. In this paper, a distributedand real-time approach for tracking multipleobjects on multiple cameras is presented. A quantitativecomparison with six state-of-the-art methods hasbeen carried out on the publicly available PETS 2009data set, demonstrating the eectiveness of the algorithm.Moreover, the proposed method has been testedalso on a multi-camera soccer data set, showing its datafusion capabilities.File in questo prodotto:
Non ci sono file associati a questo prodotto.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
