Rao-Blackwellized Resampling Particle Filter for Real-time Player Tracking in Sports

Tracking multiple targets with similiar appearance is a common task in computer vision applications, especially in sports games. We propose a Rao-Blackwellized Resampling Particle Filter (RBRPF) as an implementable real-time continuation of a state-of-the-art multi-target tracking method. Target configurations are tracked by sampling associations and solving single-target tracking problems by Kalman filters. As an advantage of the new method the independence assumption between data associations is relaxed to increase the robustness in the sports domain. Smart resampling and memoization is introduced to equip the tracking method with real-time capabilities in the first place. The probabilistic framework allows for consideration of appearance models and the fusion of different sensors. We demonstrate its applicability to real world applications by tracking soccer players captured by multiple cameras through occlusions in real-time.

[1]  Frank Dellaert,et al.  MCMC Data Association and Sparse Factorization Updating for Real Time Multitarget Tracking with Merged and Multiple Measurements , 2006, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[2]  Adrian Hilton,et al.  Football player tracking for video annotation , 2006 .

[3]  Jia Liu,et al.  Automatic Player Detection, Labeling and Tracking in Broadcast Soccer Video , 2007, BMVC.

[4]  Stefan Carlsson,et al.  Tracking and Labelling of Interacting Multiple Targets , 2006, ECCV.

[5]  Larry S. Davis,et al.  Fast multiple object tracking via a hierarchical particle filter , 2005, Tenth IEEE International Conference on Computer Vision (ICCV'05) Volume 1.

[6]  Stefan Carlsson,et al.  Multi-Target Tracking - Linking Identities using Bayesian Network Inference , 2006, 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'06).

[7]  Jean-Bernard Hayet,et al.  Collaborative Multi-Camera Tracking of Athletes in Team Sports , 2006 .

[8]  A. Doucet.  On sequential Monte Carlo methods for Bayesian filtering , 1998 .

[9]  Michael Beetz,et al.  Visually Tracking Football Games Based on TV Broadcasts , 2007, IJCAI.

[10]  Michael Beetz,et al.  An Adaptive Vision System for Tracking Soccer Players from Variable Camera Settings , 2007, ICVS 2007.

[11]  Justus H. Piater,et al.  Multi-camera People Tracking by Collaborative Particle Filters and Principal Axis-Based Integration , 2007, ACCV.

[12]  Michael Beetz,et al.  Camera-based observation of football games for analyzing multi-agent activities , 2006, AAMAS '06.

[13]  Andrew Blake,et al.  A Probabilistic Exclusion Principle for Tracking Multiple Objects , 2000, Proceedings of the Seventh IEEE International Conference on Computer Vision.

[14]  Jouko Lampinen,et al.  Rao-Blackwellized particle filter for multiple target tracking , 2007, Inf. Fusion.

[15]  Thia Kirubarajan,et al.  Estimation with Applications to Tracking and Navigation: Theory, Algorithms and Software , 2001 .

[16]  Ricardo M. L. Barros,et al.  Tracking soccer players aiming their kinematical motion analysis , 2006, Comput. Vis. Image Underst..

[17]  Oliver Grau,et al.  Tracking football player movement from a single moving camera using particle filters , 2006 .

[18]  Neil J. Gordon,et al.  A tutorial on particle filters for online nonlinear/non-Gaussian Bayesian tracking , 2002, IEEE Trans. Signal Process..

[19]  Yan Li,et al.  Evaluating the performance of systems for tracking football players and ball , 2005, IEEE Conference on Advanced Video and Signal Based Surveillance, 2005..

[20]  John R. Kender,et al.  Robust Methods and Representations for Soccer Player Tracking and Collision Resolution , 2005, CIVR.

[21]  Jouko Lampinen,et al.  Association for Multiple Target Tracking , 2004 .