Multi-target Tracking on a Large Scale: Experiences from Football Player Tracking
作者： P. Nillius , Stefan Carlsson , J. Sullivan
DOI:
关键词:
摘要: Multi-target tracking requires locating the targets and labeling their identities. The latter is a challenge when many targets, with indistinct appearances, frequently occlude one another, as in football surveillance tracking. We present an approach to solving this problem. When isolated, target can be tracked its identity maintained. While, if interact not always case. build track graph which denotes are isolated describes how they interact. Measures of similarity between tracks defined. goal associate identities tracks, by exploiting constraints measures. formulate Bayesian network inference problem, allowing us use standard message propagation find most probable set paths efficient way. high complexity inevitable large problems gracefully reduced removing dependency links tracks. apply method 10 min sequence international game compare results ground truth.
参考文章(12)
Zia Khan, Tucker Balch, Frank Dellaert, An MCMC-Based Particle Filter for Tracking Multiple Interacting Targets european conference on computer vision. pp. 279- 290 ,(2004) , 10.1007/978-3-540-24673-2_23
Robert G. Cowell, V. Nair, David J. Spiegelhalter, Steffen L. Lauritzen, A. Philip David, M. Jordan, J. Lawless, Probabilistic Networks and Expert Systems In: UNSPECIFIED Springer-Verlag (1999). ,(1999)
Ming Xu, James Orwell, Graeme Jones, Tracking football players with multiple cameras international conference on image processing. ,vol. 5, pp. 2909- 2912 ,(2004) , 10.1049/PBPC005E_CH8
G. Medioni, R.M.L. Barros, P. Figueroa, N. Leite, I. Cohen, Tracking soccer players using the graph representation international conference on pattern recognition. ,vol. 4, pp. 787- 790 ,(2004) , 10.1109/ICPR.2004.950
Cecil Huang, Adnan Darwiche, Inference in belief networks: A procedural guide International Journal of Approximate Reasoning. ,vol. 15, pp. 225- 263 ,(1996) , 10.1016/S0888-613X(96)00069-2
C J Needham, R D Boyle, Tracking multiple sports players through occlusion, congestion and scale british machine vision conference. pp. 1- 10 ,(2001) , 10.5244/C.15.11
Marc Gelgon, Patrick Bouthemy, Jean-Pierre Le Cadre, Recovery of the trajectories of multiple moving objects in an image sequence with a PMHT approach Image and Vision Computing. ,vol. 23, pp. 19- 31 ,(2005) , 10.1016/J.IMAVIS.2004.07.004
Arnaud Doucet, Jaco Vermaak, Patrick Pérez, Maintaining Multi-Modality through Mixture Tracking international conference on computer vision. pp. 1110- 1116 ,(2003)
Kenji Okuma, Ali Taleghani, Nando de Freitas, James J. Little, David G. Lowe, A Boosted Particle Filter: Multitarget Detection and Tracking european conference on computer vision. ,vol. 3021, pp. 28- 39 ,(2004) , 10.1007/978-3-540-24670-1_3
H. Saito, S. Iwase, Parallel tracking of all soccer players by integrating detected positions in multiple view images international conference on pattern recognition. ,vol. 4, pp. 751- 754 ,(2004) , 10.1109/ICPR.2004.675