An Improved Struck Tracking Method Based on Fast Search Method
- DOI
- 10.2991/lemcs-15.2015.293How to use a DOI?
- Keywords
- Struck; SVM; Fast searching method; NTSS; Object tracking
- Abstract
Traditional methods such as adaptive tracking-by-detection approaches generate a set of samples and, depending on the type of learner, producing training labels. ,However, it is not clear how to best perform their sample step. Furthermore, the objective for the classifier (label prediction) is not explicitly coupled to the objective for the tracker (accurate estimation of object position.). Then a new method named ,Struck, it avoids these steps and operates directly on the tracking .output. Struck uses a kernelized structured output support vector machine (SVM), which is learned ,online to provide adaptive tracking. What’s ,more, to allow for real-time application, it applies a budgeting mechanism which prevents the unbounded growth in the number of support vectors which would otherwise occur during tracking. However, this method does not run fast and may affect its real-time performance. To further improve its operating speed and simplify algorithm without reducing much ,accuracy, researchers introduce fast searching method to replace its original initial sampling and change some of its default .parameters. Just from the amount of ,calculation, the method can partly develop algorithm speed with good performance.
- Copyright
- © 2015, the Authors. Published by Atlantis Press.
- Open Access
- This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).
Cite this article
TY - CONF AU - Yunfeng Ni AU - Li Xu AU - Ying Hou PY - 2015/07 DA - 2015/07 TI - An Improved Struck Tracking Method Based on Fast Search Method BT - Proceedings of the International Conference on Logistics, Engineering, Management and Computer Science PB - Atlantis Press SP - 1473 EP - 1478 SN - 1951-6851 UR - https://doi.org/10.2991/lemcs-15.2015.293 DO - 10.2991/lemcs-15.2015.293 ID - Ni2015/07 ER -