Robust visual tracking via two-stage binocular sparse learning

The Journal of Engineering

Ziang Ma¹  Wei Lu² 

[1] Zhejiang Dahua Technology CO., LTD. , Zhejiang Province , Hangzhou , People's Republic of China

关键词: two-stage sparse representation-based method;    robustness;    enforcing joint sparsity;    robust visual tracking;    novel stereo vision;    robust feature-level fusion;    depth view;    colouring information-based features;    two-stage binocular sparse learning;    sparse optimisation;    robust representation;    robust tracking;    pruned views;    unreliable features;    low rank constraint;    target object;    multiple features;    depth-based histogram analysis;    objective function;    appearance modelling;   

DOI  :  10.1049/joe.2018.8328

学科分类：工程和技术（综合）

来源: IET

【 摘 要 】

Combining multiple features and enforcing joint sparsity have proven to be beneficial for robust tracking. In this study, a novel stereo vision and two-stage sparse representation-based method is presented. First, the colouring information-based features are augmented with a depth view in the appearance modelling of a target object. Unreliable features are then dynamically removed for robust feature-level fusion in the first stage of sparse optimisation. Next, the low rank constraint is imposed onto the objective function, which facilitates a more robust representation of the ensemble of particles over the pruned views. Finally, the authors propose to detect occlusion via depth-based histogram analysis to guarantee the effectiveness of the template update. Experiments are performed on two large-scale benchmark datasets: KITTI and Princeton. Authors’ approach achieves state-of-the-art results in the aspect of robustness and accuracy.

【 授权许可】

CC BY