Abstract:Multiple feature object representation has been proved as a robust approach for visual tracking. Different types of situations such as occlusion, rotation and illumination may occur during tracking, especially long sequences. Robust tracking could be obtained as multiple features could complement each other. In this paper, we cast visual tracking as a novel multi-task sparse learning problem and exploit various types of visual features, such as intensity, color, texture and edge, where each feature can be sparsely represented by a linear combination of atoms from an adaptive feature template. We use an on-line feature selection mechanism based on the two-class variance ratio measure, applied to log likelihood distributions computed with respect to a given feature from samples of object and background pixels. The proposed method is integrated in a particle filtering framework. We jointly consider the underlying relationship across different particles, and tackle it in a unified robust multi-task formulation. In addition, to capture the frequently emerging outlier tasks, we make fully use of a decomposition model which enables a more robust and accurate approximation. We show that the proposed model can be efficiently solved using the Alternating direction method of multipliers (ADMM) with a small number of closed-form updates. Four types of features are implemented and tested on numerous benchmark video sequences. Both the qualitative and quantitative results demonstrate the superior performance of the proposed approach compared to 9 state of-the-art trackers.

On Combining Compressed Sensing and Sparse Representations for Object Tracking.

Supplementary Material: Quasi-Dense Similarity Learning for Multiple Object Tracking

Object tracking with 3D LIDAR via multi-task sparse learning

Adaptive And Compressive Target Tracking Based On Feature Point Matching

Tracking in Multimedia Data Via Robust Reweighted Local Multi-Task Sparse Representation for Transportation Surveillance

Real-time Visual Tracking Using Sparse Representation

Exploiting More Information in Sparse Point Cloud for 3D Single Object Tracking

Object Tracking Via Appearance Modeling and Sparse Representation

Object Tracking by Occlusion Detection via Structured Sparse Learning

Visual tracking via robust multi-task multi-feature joint sparse representation

Object Tracking with Joint Optimization of Representation and Classification

Real-Time Visual Tracking Using Compressive Sensing

SparseTrack: Multi-Object Tracking by Performing Scene Decomposition based on Pseudo-Depth

Beyond Local Search: Tracking Objects Everywhere with Instance-Specific Proposals

Spatial Neighborhood-Constrained Linear Coding for Visual Object Tracking

Tracking via Robust Multi-task Multi-view Joint Sparse Representation

Weighted sparse coding residual minimization for visual tracking

Robust Structural Sparse Tracking

Robust object tracking via multi-scale patch based sparse coding histogram

Visual Tracking via Sparse Representation and Online Dictionary Learning.

Visual Tracking Using Iterative Sparse Approximation