Real-time Visual Tracking Using Sparse Representation

TL;DR

This paper introduces a real-time compressed sensing-based tracking algorithm that significantly accelerates sparse representation tracking while maintaining high accuracy, and further improves performance with a background model for stationary cameras.

Contribution

The paper presents a novel real-time tracking method using compressed sensing and a customized OMP algorithm, achieving up to 6,000 times faster speed with competitive accuracy.

Findings

Achieves real-time tracking up to 6,000 times faster than traditional $$ tracker.

Maintains or exceeds the accuracy of existing $$ trackers.

Outperforms state-of-the-art methods with the CS-based background model in stationary camera scenarios.

Abstract

The $\ell_1$ tracker obtains robustness by seeking a sparse representation of the tracking object via $\ell_1$ norm minimization \cite{Xue_ICCV_09_Track}. However, the high computational complexity involved in the $ \ell_1 $ tracker restricts its further applications in real time processing scenario. Hence we propose a Real Time Compressed Sensing Tracking (RTCST) by exploiting the signal recovery power of Compressed Sensing (CS). Dimensionality reduction and a customized Orthogonal Matching Pursuit (OMP) algorithm are adopted to accelerate the CS tracking. As a result, our algorithm achieves a real-time speed that is up to $6,000$ times faster than that of the $\ell_1$ tracker. Meanwhile, RTCST still produces competitive (sometimes even superior) tracking accuracy comparing to the existing $\ell_1$ tracker. Furthermore, for a stationary camera, a further refined tracker is designed by integrating a CS-based background model (CSBM). This CSBM-equipped tracker coined as RTCST-B, outperforms most state-of-the-arts with respect to both accuracy and robustness. Finally, our experimental results on various video sequences, which are verified by a new metric---Tracking Success Probability (TSP), show the excellence of the proposed algorithms.