L0-Regularized Object Representation for Visual Tracking

Abstract

In this paper, we propose a robust visual tracking method by L0-regularized prior in a particle filter framework. In contrast to existing methods, the proposed method employs L0 norm to regularize the linear coefficients of incrementally updated linear basis. The sparsity constraint enables the tracker to effectively handle difficult challenges, such as occlusion or image corruption. To achieve realtime processing, we propose a fast and efficient numerical algorithm for solving the proposed L0-regularized model. Although it is an NP-hard problem, the proposed accelerated proximal gradient (APG) approach is guaranteed to converge to a solution quickly. Extensive experimental results on challenging video sequences demonstrate that the proposed method achieves state-of-the-art results both in accuracy and speed.