Visual Tracking under Motion Blur

Bo Ma, Lianghua Huang, Jianbing Shen, Ling Shao, Ming Hsuan Yang, Fatih Porikli

Research output: Contribution to journalArticle

39 Citations (Scopus)

Abstract

Most existing tracking algorithms do not explicitly consider the motion blur contained in video sequences, which degrades their performance in real-world applications where motion blur often occurs. In this paper, we propose to solve the motion blur problem in visual tracking in a unified framework. Specifically, a joint blur state estimation and multi-task reverse sparse learning framework are presented, where the closed-form solution of blur kernel and sparse code matrix is obtained simultaneously. The reverse process considers the blurry candidates as dictionary elements, and sparsely represents blurred templates with the candidates. By utilizing the information contained in the sparse code matrix, an efficient likelihood model is further developed, which quickly excludes irrelevant candidates and narrows the particle scale down. Experimental results on the challenging benchmarks show that our method performs well against the state-of-the-art trackers.

Original languageEnglish
Article number7585089
Pages (from-to)5867-5876
Number of pages10
JournalIEEE Transactions on Image Processing
Volume25
Issue number12
DOIs
Publication statusPublished - 2016 Dec

Fingerprint

State estimation
Glossaries

All Science Journal Classification (ASJC) codes

  • Software
  • Computer Graphics and Computer-Aided Design

Cite this

Ma, B., Huang, L., Shen, J., Shao, L., Yang, M. H., & Porikli, F. (2016). Visual Tracking under Motion Blur. IEEE Transactions on Image Processing, 25(12), 5867-5876. [7585089]. https://doi.org/10.1109/TIP.2016.2615812
Ma, Bo ; Huang, Lianghua ; Shen, Jianbing ; Shao, Ling ; Yang, Ming Hsuan ; Porikli, Fatih. / Visual Tracking under Motion Blur. In: IEEE Transactions on Image Processing. 2016 ; Vol. 25, No. 12. pp. 5867-5876.
@article{77077c4e5b044d019cf016946976192c,
title = "Visual Tracking under Motion Blur",
abstract = "Most existing tracking algorithms do not explicitly consider the motion blur contained in video sequences, which degrades their performance in real-world applications where motion blur often occurs. In this paper, we propose to solve the motion blur problem in visual tracking in a unified framework. Specifically, a joint blur state estimation and multi-task reverse sparse learning framework are presented, where the closed-form solution of blur kernel and sparse code matrix is obtained simultaneously. The reverse process considers the blurry candidates as dictionary elements, and sparsely represents blurred templates with the candidates. By utilizing the information contained in the sparse code matrix, an efficient likelihood model is further developed, which quickly excludes irrelevant candidates and narrows the particle scale down. Experimental results on the challenging benchmarks show that our method performs well against the state-of-the-art trackers.",
author = "Bo Ma and Lianghua Huang and Jianbing Shen and Ling Shao and Yang, {Ming Hsuan} and Fatih Porikli",
year = "2016",
month = "12",
doi = "10.1109/TIP.2016.2615812",
language = "English",
volume = "25",
pages = "5867--5876",
journal = "IEEE Transactions on Image Processing",
issn = "1057-7149",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "12",

}

Ma, B, Huang, L, Shen, J, Shao, L, Yang, MH & Porikli, F 2016, 'Visual Tracking under Motion Blur', IEEE Transactions on Image Processing, vol. 25, no. 12, 7585089, pp. 5867-5876. https://doi.org/10.1109/TIP.2016.2615812

Visual Tracking under Motion Blur. / Ma, Bo; Huang, Lianghua; Shen, Jianbing; Shao, Ling; Yang, Ming Hsuan; Porikli, Fatih.

In: IEEE Transactions on Image Processing, Vol. 25, No. 12, 7585089, 12.2016, p. 5867-5876.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Visual Tracking under Motion Blur

AU - Ma, Bo

AU - Huang, Lianghua

AU - Shen, Jianbing

AU - Shao, Ling

AU - Yang, Ming Hsuan

AU - Porikli, Fatih

PY - 2016/12

Y1 - 2016/12

N2 - Most existing tracking algorithms do not explicitly consider the motion blur contained in video sequences, which degrades their performance in real-world applications where motion blur often occurs. In this paper, we propose to solve the motion blur problem in visual tracking in a unified framework. Specifically, a joint blur state estimation and multi-task reverse sparse learning framework are presented, where the closed-form solution of blur kernel and sparse code matrix is obtained simultaneously. The reverse process considers the blurry candidates as dictionary elements, and sparsely represents blurred templates with the candidates. By utilizing the information contained in the sparse code matrix, an efficient likelihood model is further developed, which quickly excludes irrelevant candidates and narrows the particle scale down. Experimental results on the challenging benchmarks show that our method performs well against the state-of-the-art trackers.

AB - Most existing tracking algorithms do not explicitly consider the motion blur contained in video sequences, which degrades their performance in real-world applications where motion blur often occurs. In this paper, we propose to solve the motion blur problem in visual tracking in a unified framework. Specifically, a joint blur state estimation and multi-task reverse sparse learning framework are presented, where the closed-form solution of blur kernel and sparse code matrix is obtained simultaneously. The reverse process considers the blurry candidates as dictionary elements, and sparsely represents blurred templates with the candidates. By utilizing the information contained in the sparse code matrix, an efficient likelihood model is further developed, which quickly excludes irrelevant candidates and narrows the particle scale down. Experimental results on the challenging benchmarks show that our method performs well against the state-of-the-art trackers.

UR - http://www.scopus.com/inward/record.url?scp=84994876750&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84994876750&partnerID=8YFLogxK

U2 - 10.1109/TIP.2016.2615812

DO - 10.1109/TIP.2016.2615812

M3 - Article

AN - SCOPUS:84994876750

VL - 25

SP - 5867

EP - 5876

JO - IEEE Transactions on Image Processing

JF - IEEE Transactions on Image Processing

SN - 1057-7149

IS - 12

M1 - 7585089

ER -

Ma B, Huang L, Shen J, Shao L, Yang MH, Porikli F. Visual Tracking under Motion Blur. IEEE Transactions on Image Processing. 2016 Dec;25(12):5867-5876. 7585089. https://doi.org/10.1109/TIP.2016.2615812