Robust Scale Adaptive and Real-Time Visual Tracking with Correlation Filters

Jiatian PI; Keli HU; Yuzhang GU; Lei QU; Fengrong LI; Xiaolin ZHANG; Yunlong ZHAN

doi:10.1587/transinf.2015EDP7459

Robust Scale Adaptive and Real-Time Visual Tracking with Correlation Filters

Jiatian PI, Keli HU, Yuzhang GU, Lei QU, Fengrong LI, Xiaolin ZHANG, Yunlong ZHAN

Full Text Views

0

Cite this

Summary :

Visual tracking has been studied for several decades but continues to draw significant attention because of its critical role in many applications. Recent years have seen greater interest in the use of correlation filters in visual tracking systems, owing to their extremely compelling results in different competitions and benchmarks. However, there is still a need to improve the overall tracking capability to counter various tracking issues, including large scale variation, occlusion, and deformation. This paper presents an appealing tracker with robust scale estimation, which can handle the problem of fixed template size in Kernelized Correlation Filter (KCF) tracker with no significant decrease in the speed. We apply the discriminative correlation filter for scale estimation as an independent part after finding the optimal translation based on the KCF tracker. Compared to an exhaustive scale space search scheme, our approach provides improved performance while being computationally efficient. In order to reveal the effectiveness of our approach, we use benchmark sequences annotated with 11 attributes to evaluate how well the tracker handles different attributes. Numerous experiments demonstrate that the proposed algorithm performs favorably against several state-of-the-art algorithms. Appealing results both in accuracy and robustness are also achieved on all 51 benchmark sequences, which proves the efficiency of our tracker.

Publication: IEICE TRANSACTIONS on Information Vol.E99-D No.7 pp.1895-1902

Publication Date: 2016/07/01

Publicized: 2016/04/07

Online ISSN: 1745-1361

DOI: 10.1587/transinf.2015EDP7459

Type of Manuscript: PAPER

Category: Image Recognition, Computer Vision

Authors

Jiatian PI
  SIMIT
Keli HU
  Shaoxing University
Yuzhang GU
  SIMIT
Lei QU
  SIMIT
Fengrong LI
  SIMIT
Xiaolin ZHANG
  SIMIT
Yunlong ZHAN
  SIMIT

Keyword

correlation filters, kernel methods, scale estimation, visual tracking

Cite this

Copy

Jiatian PI, Keli HU, Yuzhang GU, Lei QU, Fengrong LI, Xiaolin ZHANG, Yunlong ZHAN, "Robust Scale Adaptive and Real-Time Visual Tracking with Correlation Filters" in IEICE TRANSACTIONS on Information, vol. E99-D, no. 7, pp. 1895-1902, July 2016, doi: 10.1587/transinf.2015EDP7459.
Abstract: Visual tracking has been studied for several decades but continues to draw significant attention because of its critical role in many applications. Recent years have seen greater interest in the use of correlation filters in visual tracking systems, owing to their extremely compelling results in different competitions and benchmarks. However, there is still a need to improve the overall tracking capability to counter various tracking issues, including large scale variation, occlusion, and deformation. This paper presents an appealing tracker with robust scale estimation, which can handle the problem of fixed template size in Kernelized Correlation Filter (KCF) tracker with no significant decrease in the speed. We apply the discriminative correlation filter for scale estimation as an independent part after finding the optimal translation based on the KCF tracker. Compared to an exhaustive scale space search scheme, our approach provides improved performance while being computationally efficient. In order to reveal the effectiveness of our approach, we use benchmark sequences annotated with 11 attributes to evaluate how well the tracker handles different attributes. Numerous experiments demonstrate that the proposed algorithm performs favorably against several state-of-the-art algorithms. Appealing results both in accuracy and robustness are also achieved on all 51 benchmark sequences, which proves the efficiency of our tracker.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2015EDP7459/_p

Copy

@ARTICLE{e99-d_7_1895,
author={Jiatian PI, Keli HU, Yuzhang GU, Lei QU, Fengrong LI, Xiaolin ZHANG, Yunlong ZHAN, },
journal={IEICE TRANSACTIONS on Information},
title={Robust Scale Adaptive and Real-Time Visual Tracking with Correlation Filters},
year={2016},
volume={E99-D},
number={7},
pages={1895-1902},
abstract={Visual tracking has been studied for several decades but continues to draw significant attention because of its critical role in many applications. Recent years have seen greater interest in the use of correlation filters in visual tracking systems, owing to their extremely compelling results in different competitions and benchmarks. However, there is still a need to improve the overall tracking capability to counter various tracking issues, including large scale variation, occlusion, and deformation. This paper presents an appealing tracker with robust scale estimation, which can handle the problem of fixed template size in Kernelized Correlation Filter (KCF) tracker with no significant decrease in the speed. We apply the discriminative correlation filter for scale estimation as an independent part after finding the optimal translation based on the KCF tracker. Compared to an exhaustive scale space search scheme, our approach provides improved performance while being computationally efficient. In order to reveal the effectiveness of our approach, we use benchmark sequences annotated with 11 attributes to evaluate how well the tracker handles different attributes. Numerous experiments demonstrate that the proposed algorithm performs favorably against several state-of-the-art algorithms. Appealing results both in accuracy and robustness are also achieved on all 51 benchmark sequences, which proves the efficiency of our tracker.},
keywords={},
doi={10.1587/transinf.2015EDP7459},
ISSN={1745-1361},
month={July},}

Copy

TY - JOUR
TI - Robust Scale Adaptive and Real-Time Visual Tracking with Correlation Filters
T2 - IEICE TRANSACTIONS on Information
SP - 1895
EP - 1902
AU - Jiatian PI
AU - Keli HU
AU - Yuzhang GU
AU - Lei QU
AU - Fengrong LI
AU - Xiaolin ZHANG
AU - Yunlong ZHAN
PY - 2016
DO - 10.1587/transinf.2015EDP7459
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E99-D
IS - 7
JA - IEICE TRANSACTIONS on Information
Y1 - July 2016
AB - Visual tracking has been studied for several decades but continues to draw significant attention because of its critical role in many applications. Recent years have seen greater interest in the use of correlation filters in visual tracking systems, owing to their extremely compelling results in different competitions and benchmarks. However, there is still a need to improve the overall tracking capability to counter various tracking issues, including large scale variation, occlusion, and deformation. This paper presents an appealing tracker with robust scale estimation, which can handle the problem of fixed template size in Kernelized Correlation Filter (KCF) tracker with no significant decrease in the speed. We apply the discriminative correlation filter for scale estimation as an independent part after finding the optimal translation based on the KCF tracker. Compared to an exhaustive scale space search scheme, our approach provides improved performance while being computationally efficient. In order to reveal the effectiveness of our approach, we use benchmark sequences annotated with 11 attributes to evaluate how well the tracker handles different attributes. Numerous experiments demonstrate that the proposed algorithm performs favorably against several state-of-the-art algorithms. Appealing results both in accuracy and robustness are also achieved on all 51 benchmark sequences, which proves the efficiency of our tracker.
ER -