Discriminative Hash Tracking With Group Sparsity

SUMMARYTracking-by-detection methods consider tracking task as a continuous detection problem applied over video frames. Modern trackingby-detection trackers have online learning ability; the update stage is essential because it determines how to modify the classifier inherent in a tracker. However, most trackers search for the target within a fixed region centered at the previous object position; thus, they lack spatiotemporal consistency. This becomes a problem when the tracker detects an incorrect object during short-term occlusion. In addition, the scale of the bounding box that contains the target object is usually assumed not to change. This assumption is unrealistic for long-term tracking, where the scale of the target varies as the distance between the target and the camera changes. The accumulation of errors resulting from these shortcomings results in the drift problem, i.e. drifting away from the target object. To resolve this problem, we present a drift-free, online learning-based tracking-by-detection method using a single static camera. We improve the latent structured support vector machine (SVM) tracker by designing a more robust tracker update step by incorporating two Kalman filter modules: the first is used to predict an adaptive search region in consideration of the object motion; the second is used to adjust the scale of the bounding box by accounting for the background model. We propose a hierarchical search strategy that combines Bhattacharyya coefficient similarity analysis and Kalman predictors. This strategy facilitates overcoming occlusion and increases tracking efficiency. We evaluate this work using publicly available videos thoroughly. Experimental results show that the proposed method outperforms the state-of-theart trackers. key words: object tracking, support vector machine (SVM), latent structured SVM, online learning, Kalman filter, hierarchical search

show abstract

“…Nevertheless, most tracking algorithms can generally be classified into two categories: generative [40], [41] and discriminative [42], [43].…”

Section: E Classification Of Object Tracking Methodsmentioning

confidence: 99%

Drift-Free Tracking Surveillance Based on Online Latent Structured SVM and Kalman Filter Modules

Chen

Zhang

2018

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

show abstract

“…First, some features are extracted to represent the target, such as intensity [10], color [20], and Haar-like features [21]. To make the target representation more efficient and effective, the dimensionality of the feature can be reduced by feature selection methods [22,23] such as in [24,25], or Hashing methods [26,27] as in [28][29][30]. Then, given the extracted features, the target can be represented holistically or locally.…”

Section: Related Workmentioning

confidence: 99%

Multi-level visual tracking with hierarchical tree structural constraint

Wang

Zhu

et al. 2016

Neurocomputing

View full text Add to dashboard Cite

“…Recently, there have been important advances in the structured sparsity methodology for collision avoidance [37], [38]; path-following and tracking [39], [40]; and visual tracking [41], [42]. In [43], a discriminative supervised hashing method was proposed for object tracking tasks. An adaptive elastic echo state network was developed for multivariate time-series prediction in [44].…”

Section: Introductionmentioning

confidence: 99%

Autonomous Tracking and State Estimation With Generalized Group Lasso

Gao

Särkkä

Claveria-Vega

et al. 2022

IEEE Trans. Cybern.

View full text Add to dashboard Cite

We address the problem of autonomous tracking and state estimation for marine vessels, autonomous vehicles, and other dynamic signals under a (structured) sparsity assumption. The aim is to improve the tracking and estimation accuracy with respect to the classical Bayesian filters and smoothers. We formulate the estimation problem as a dynamic generalized group Lasso problem and develop a class of smoothing-andsplitting methods to solve it. The Levenberg-Marquardt iterated extended Kalman smoother-based multiblock alternating direction method of multipliers (LM-IEKS-mADMMs) algorithms are based on the alternating direction method of multipliers (ADMMs) framework. This leads to minimization subproblems with an inherent structure to which three new augmented recursive smoothers are applied. Our methods can deal with large-scale problems without preprocessing for dimensionality reduction. Moreover, the methods allow one to solve nonsmooth nonconvex optimization problems. We then prove that under mild conditions, the proposed methods converge to a stationary point of the optimization problem. By simulated and real-data experiments, including multisensor range measurement problems, marine vessel tracking, autonomous vehicle tracking, and audio signal restoration, we show the practical effectiveness of the proposed methods.

show abstract

Discriminative Hash Tracking With Group Sparsity

Cited by 17 publications

References 41 publications

Drift-Free Tracking Surveillance Based on Online Latent Structured SVM and Kalman Filter Modules

Drift-Free Tracking Surveillance Based on Online Latent Structured SVM and Kalman Filter Modules

Multi-level visual tracking with hierarchical tree structural constraint

Autonomous Tracking and State Estimation With Generalized Group Lasso

Contact Info

Product

Resources

About