Object tracking via dense SIFT features and low-rank representation

In this paper, we present a low-rank sparse tracking method which builds upon the particle filtering framework. The proposed method learns the local dense scale-invariant feature transform features corresponding to candidate samples jointly by exploiting the underlying sparse and low-rank constraint...

Full description

Saved in:
Bibliographic Details
Published in:Soft computing (Berlin, Germany) Germany), 2019-10, Vol.23 (20), p.10173-10186
Main Authors: Wang, Yong, Luo, Xinbin, Ding, Lu, Wu, Jingjing
Format: Article
Language:English
Subjects:
Algorithms
Artificial Intelligence
Computational Intelligence
Control
Decomposition
Engineering
Feature extraction
Mathematical Logic and Foundations
Mechatronics
Methodologies and Application
Methods
Robotics
Sparsity
Tracking
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In this paper, we present a low-rank sparse tracking method which builds upon the particle filtering framework. The proposed method learns the local dense scale-invariant feature transform features corresponding to candidate samples jointly by exploiting the underlying sparse and low-rank constraints. Furthermore, the alternating direction method of multipliers method guarantees the optimization equation can be solved accurately and robustly. We evaluate our proposed tracking method against 9 state-of-the-art trackers on a set of 64 challenging sequences. Experimental results show that the proposed method performs favorably against state-of-the-art trackers in terms of accuracy.
ISSN:1432-7643
1433-7479
DOI:10.1007/s00500-018-3571-5