Robust superpixel tracking

Fan Yang, Huchuan Lu, Ming Hsuan Yang

Research output: Contribution to journalArticle

234 Citations (Scopus)

Abstract

While numerous algorithms have been proposed for object tracking with demonstrated success, it remains a challenging problem for a tracker to handle large appearance change due to factors such as scale, motion, shape deformation, and occlusion. One of the main reasons is the lack of effective image representation schemes to account for appearance variation. Most of the trackers use high-level appearance structure or low-level cues for representing and matching target objects. In this paper, we propose a tracking method from the perspective of midlevel vision with structural information captured in superpixels. We present a discriminative appearance model based on superpixels, thereby facilitating a tracker to distinguish the target and the background with midlevel cues. The tracking task is then formulated by computing a target-background confidence map, and obtaining the best candidate by maximum a posterior estimate. Experimental results demonstrate that our tracker is able to handle heavy occlusion and recover from drifts. In conjunction with online update, the proposed algorithm is shown to perform favorably against existing methods for object tracking. Furthermore, the proposed algorithm facilitates foreground and background segmentation during tracking.

Original languageEnglish
Article number6718099
Pages (from-to)1639-1651
Number of pages13
JournalIEEE Transactions on Image Processing
Volume23
Issue number4
DOIs
Publication statusPublished - 2014 Apr

All Science Journal Classification (ASJC) codes

  • Software
  • Computer Graphics and Computer-Aided Design

Cite this

Yang, Fan ; Lu, Huchuan ; Yang, Ming Hsuan. / Robust superpixel tracking. In: IEEE Transactions on Image Processing. 2014 ; Vol. 23, No. 4. pp. 1639-1651.
@article{a715b1e9a9124522b1543df3a8bbea37,
title = "Robust superpixel tracking",
abstract = "While numerous algorithms have been proposed for object tracking with demonstrated success, it remains a challenging problem for a tracker to handle large appearance change due to factors such as scale, motion, shape deformation, and occlusion. One of the main reasons is the lack of effective image representation schemes to account for appearance variation. Most of the trackers use high-level appearance structure or low-level cues for representing and matching target objects. In this paper, we propose a tracking method from the perspective of midlevel vision with structural information captured in superpixels. We present a discriminative appearance model based on superpixels, thereby facilitating a tracker to distinguish the target and the background with midlevel cues. The tracking task is then formulated by computing a target-background confidence map, and obtaining the best candidate by maximum a posterior estimate. Experimental results demonstrate that our tracker is able to handle heavy occlusion and recover from drifts. In conjunction with online update, the proposed algorithm is shown to perform favorably against existing methods for object tracking. Furthermore, the proposed algorithm facilitates foreground and background segmentation during tracking.",
author = "Fan Yang and Huchuan Lu and Yang, {Ming Hsuan}",
year = "2014",
month = "4",
doi = "10.1109/TIP.2014.2300823",
language = "English",
volume = "23",
pages = "1639--1651",
journal = "IEEE Transactions on Image Processing",
issn = "1057-7149",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "4",

}

Robust superpixel tracking. / Yang, Fan; Lu, Huchuan; Yang, Ming Hsuan.

In: IEEE Transactions on Image Processing, Vol. 23, No. 4, 6718099, 04.2014, p. 1639-1651.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Robust superpixel tracking

AU - Yang, Fan

AU - Lu, Huchuan

AU - Yang, Ming Hsuan

PY - 2014/4

Y1 - 2014/4

N2 - While numerous algorithms have been proposed for object tracking with demonstrated success, it remains a challenging problem for a tracker to handle large appearance change due to factors such as scale, motion, shape deformation, and occlusion. One of the main reasons is the lack of effective image representation schemes to account for appearance variation. Most of the trackers use high-level appearance structure or low-level cues for representing and matching target objects. In this paper, we propose a tracking method from the perspective of midlevel vision with structural information captured in superpixels. We present a discriminative appearance model based on superpixels, thereby facilitating a tracker to distinguish the target and the background with midlevel cues. The tracking task is then formulated by computing a target-background confidence map, and obtaining the best candidate by maximum a posterior estimate. Experimental results demonstrate that our tracker is able to handle heavy occlusion and recover from drifts. In conjunction with online update, the proposed algorithm is shown to perform favorably against existing methods for object tracking. Furthermore, the proposed algorithm facilitates foreground and background segmentation during tracking.

AB - While numerous algorithms have been proposed for object tracking with demonstrated success, it remains a challenging problem for a tracker to handle large appearance change due to factors such as scale, motion, shape deformation, and occlusion. One of the main reasons is the lack of effective image representation schemes to account for appearance variation. Most of the trackers use high-level appearance structure or low-level cues for representing and matching target objects. In this paper, we propose a tracking method from the perspective of midlevel vision with structural information captured in superpixels. We present a discriminative appearance model based on superpixels, thereby facilitating a tracker to distinguish the target and the background with midlevel cues. The tracking task is then formulated by computing a target-background confidence map, and obtaining the best candidate by maximum a posterior estimate. Experimental results demonstrate that our tracker is able to handle heavy occlusion and recover from drifts. In conjunction with online update, the proposed algorithm is shown to perform favorably against existing methods for object tracking. Furthermore, the proposed algorithm facilitates foreground and background segmentation during tracking.

UR - http://www.scopus.com/inward/record.url?scp=84896128090&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84896128090&partnerID=8YFLogxK

U2 - 10.1109/TIP.2014.2300823

DO - 10.1109/TIP.2014.2300823

M3 - Article

C2 - 24808336

AN - SCOPUS:84896128090

VL - 23

SP - 1639

EP - 1651

JO - IEEE Transactions on Image Processing

JF - IEEE Transactions on Image Processing

SN - 1057-7149

IS - 4

M1 - 6718099

ER -