Tracking across multiple cameras with disjoint views
- 1 January 2003
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 952-957 vol.2
- https://doi.org/10.1109/iccv.2003.1238451
Abstract
Conventional tracking approaches assume proximity in space, time and appearance of objects in successive observations. However, observations of objects are often widely separated in time and space when viewed from multiple non-overlapping cameras. To address this problem, we present a novel approach for establishing object correspondence across non-overlapping cameras. Our multicamera tracking algorithm exploits the redundance in paths that people and cars tend to follow, e.g. roads, walk-ways or corridors, by using motion trends and appearance of objects, to establish correspondence. Our system does not require any inter-camera calibration, instead the system learns the camera topology and path probabilities of objects using Parzen windows, during a training phase. Once the training is complete, correspondences are assigned using the maximum a posteriori (MAP) estimation framework. The learned parameters are updated with changing trajectory patterns. Experiments with real world videos are reported, which validate the proposed approach.Keywords
This publication has 11 references indexed in Scilit:
- Bayesian multi-camera surveillancePublished by Institute of Electrical and Electronics Engineers (IEEE) ,2003
- A non-iterative greedy algorithm for multi-frame point correspondencePublished by Institute of Electrical and Electronics Engineers (IEEE) ,2003
- Tracking multiple people with a multi-camera systemPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Multiple camera fusion for multi-object trackingPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Human tracking in multiple camerasPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Algorithms for cooperative multisensor surveillanceProceedings of the IEEE, 2001
- Monitoring activities from multiple video streams: establishing a common coordinate framePublished by Institute of Electrical and Electronics Engineers (IEEE) ,2000
- Efficient On-Line Nonparametric Kernel Density EstimationAlgorithmica, 1999
- Tracking human motion in structured environments using a distributed-camera systemPublished by Institute of Electrical and Electronics Engineers (IEEE) ,1999
- An $n^{5/2} $ Algorithm for Maximum Matchings in Bipartite GraphsSIAM Journal on Computing, 1973