.png){kind=icon}
.png){kind=icon}
Return
Kernel joint visual tracking and recognition based on structured sparse representation
.png){kind=icon}
PRE
AI
DOI：10.1016/j.neucom.2016.02.024.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Visual target tracking and recognition are closely interrelated and can help each other potentially and significantly. The optimal Bayes joint decision and estimation (JDE) model guarantees the general decision (recognition) and estimation (tracking) arriving at the global optimization. In this paper, a kernel joint tracking and recognition (KJTR) algorithm is proposed to apply JDE to the simultaneous visual tracking and recognition problem and use the iteration learning procedure to achieve optimization. Besides, the structured sparse representation (SSR) model shows great efficiency and robustness in exploiting both holistic and local information of the target appearance. We show that constructing the appearance model with SSR can improve the performance of the proposed algorithm. Then, the contribution of each test candidate is considered into the learning procedure by a kernel function. Furthermore, the new joint weights of KJTR provide flexibility with appearance changes and thus robustness to the dynamic scene. The experimental results demonstrate that the KJTR algorithm performs well in terms of accuracy and robustness. (C) 2016 Elsevier B.V. All rights reserved.
Keywords:
Joint visual tracking and recognition
Joint decision and estimation
Structured sparse representation
Kernel function
Joint weight
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.