Robust Visual Tracking via Inverse Nonnegative Matrix Factorization

TL;DR

This paper introduces an inverse nonnegative matrix factorization approach for robust visual tracking, leveraging foreground and background information with sparse basis matrices, and uses SVM for classification.

Contribution

It proposes a novel inverse NMF method that enhances appearance modeling by incorporating background info and local constraints, improving tracking robustness.

Findings

Outperforms seven state-of-the-art methods in tests.

Demonstrates robustness in various video scenarios.

Effective separation of target from background.

Abstract

The establishment of robust target appearance model over time is an overriding concern in visual tracking. In this paper, we propose an inverse nonnegative matrix factorization (NMF) method for robust appearance modeling. Rather than using a linear combination of nonnegative basis matrices for each target image patch in the conventional NMF, the proposed method is a reverse thought to conventional NMF tracker. It utilizes both the foreground and background information, and imposes a local coordinate constraint, where the basis matrix is sparse matrix from the linear combination of candidates with corresponding nonnegative coefficient vectors. Inverse NMF is used as a feature encoder, where the resulting coefficient vectors are fed into a SVM classifier for separating the target from the background. The proposed method is tested on several videos and compared with seven state-of-the-art methods. Our results have provided further support to the effectiveness and robustness of the proposed method.