Robust objectness tracking with weighted multiple instance learning algorithm

A novel improved online weighted multiple instance learning algorithm(IWMIL) for visual tracking is proposed. In the IWMIL algorithm, the importance of each sample contributing to bag probability is evaluated based on the objectness estimation with object properties (superpixel straddling). To reduc...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Neurocomputing (Amsterdam) Ročník 288; s. 43 - 53
Hlavní autoři: Yang, Honghong, Qu, Shiru, Zhu, Fumin, Zheng, Zunxin
Médium: Journal Article
Jazyk:angličtina
Vydáno: Elsevier B.V 02.05.2018
Témata:
Adaptive learning rate
Object similarity constraint
Object tracking
Objectness
Weighted multiple instance learning algorithm
Weighted multiple instance learning algorithm
Objectness
Adaptive learning rate
Object tracking
Object similarity constraint
ISSN:0925-2312, 1872-8286
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:A novel improved online weighted multiple instance learning algorithm(IWMIL) for visual tracking is proposed. In the IWMIL algorithm, the importance of each sample contributing to bag probability is evaluated based on the objectness estimation with object properties (superpixel straddling). To reduce the computation cost, a coarse-to-fine sample detection method is employed to detect sample for a new arriving frame. Then, an adaptive learning rate, which exploits the maximum classifier score to assign different weights to tracking result and template, is presented to update the classifiers. Furthermore, an object similarity constraint strategy is used to estimate tracking drift. Experimental results on challenging sequences show that the proposed method is robust to occlusion and appearance changes.
ISSN:0925-2312
1872-8286
DOI:10.1016/j.neucom.2017.02.106