CMP logo

Tracking by an Optimal Sequence of Linear Predictors

Karel Zimmermann, Tomas Svoboda, Jiri Matas 		Center for Machine Perception
Czech Technical University Prague
http://cmp.felk.cvut.cz/

Movies

Motion blur (images)

Download

Abstract

We propose a learnable tracking method, where the computational complexity is explictly taken into account. The computational complexity, defined as the total number of pixels used for linear motion prediction, is minimized during an offline learning stage. Learning algorithm, based on linear and dynamic programming, searches for an optimal predictor, given a range of admissible motions, desired accuracy and a set of training examples. As a result, the globally optimal sequence of predictors covering the object is delivered. The local motion is determined only by a few hundreds of multiplications resulting in tracking with a fraction of the processing power of a desktop PC. The global motion of the object is robustly estimated by Ransac algorithm from some local motions. We also make explicit the trade-off between the number of estimated local motions and the number of Ransac iterations. The optimal predictor is applied to real-time object tracking and its performance superiority and robustness are experimentally demonstrated.

Support

References

[1] K. Zimmermann, J. Matas, and T. Svoboda. Tracking by an Optimal Sequence of Linear Predictors. IEEE Transactions on Pattern Analysis and Machine Intelligence. 31(4), 2009, [pdf]
[2] K. Zimmermann, T. Svoboda, J. Matas. Simultaneous learning of motion and appearance. The 1st International Workshop on Machine Learning for Vision-based Motion Analysis, In conjunction with the 10th European Conference on Computer Vision 2008, [ paper | talk ].
[3] K. Zimmernann, T. Svoboda, J. Matas. Adaptive Parameter Optimization for Real-time Tracking. In Proceedings of the Workshop on Non-rigid Registration and Tracking through Learning - NRTL 2007 (joint workshop with the ICCV 2007). [ paper ]
[4] J.Matas, K.Zimmermann, T. Svoboda, A. Hilton, Learning Efficient Linear Predictors for Motion Estimation, Proceedings of 5th Indian Conference on Computer Vision, Graphics and Image Processing, issn 0302-9743, isbn 978-3-540-68301-8, India, 2006
[5] K. Zimmermann. Fast Learnabe Methods for Object Tracking. PhD Thesis, Czech Technical Univeristy in Prague. [PDF]. Defended 7 Nov 2008.

MultiCam