Visual Tracking via Dynamic Graph Learning

Chenglong Li, Liang Lin, Wangmeng Zuo, Jin Tang, Ming Hsuan Yang

Research output: Contribution to journalArticle

13 Citations (Scopus)

Abstract

Existing visual tracking methods usually localize a target object with a bounding box, in which the performance of the foreground object trackers or detectors is often affected by the inclusion of background clutter. To handle this problem, we learn a patch-based graph representation for visual tracking. The tracked object is modeled by with a graph by taking a set of non-overlapping image patches as nodes, in which the weight of each node indicates how likely it belongs to the foreground and edges are weighted for indicating the appearance compatibility of two neighboring nodes. This graph is dynamically learned and applied in object tracking and model updating. The proposed algorithm performs three main steps in each frame. First, the graph is initialized by assigning binary weights of some image patches to indicate the object and background patches according to the predicted bounding box. Second, the graph is optimized to refine the patch weights. Third, the object feature representation is updated by imposing the weights of patches on the extracted image features. The object location is predicted by maximizing the classification score in the structured support vector machine. Extensive experiments show that the proposed tracking algorithm performs well against the state-of-the-art methods.

Original languageEnglish
JournalIEEE transactions on pattern analysis and machine intelligence
DOIs
Publication statusAccepted/In press - 2018 Aug 10

Fingerprint

Dynamic Graphs
Visual Tracking
Patch
Support vector machines
Graph in graph theory
Detectors
Vertex of a graph
Model Updating
Graph Representation
Experiments
Object Model
Object Tracking
Clutter
Compatibility
Object
Learning
Support Vector Machine
Inclusion
Likely
Detector

All Science Journal Classification (ASJC) codes

  • Software
  • Computer Vision and Pattern Recognition
  • Computational Theory and Mathematics
  • Artificial Intelligence
  • Applied Mathematics

Cite this

@article{617704a9173248eeb68476610612af0e,
title = "Visual Tracking via Dynamic Graph Learning",
abstract = "Existing visual tracking methods usually localize a target object with a bounding box, in which the performance of the foreground object trackers or detectors is often affected by the inclusion of background clutter. To handle this problem, we learn a patch-based graph representation for visual tracking. The tracked object is modeled by with a graph by taking a set of non-overlapping image patches as nodes, in which the weight of each node indicates how likely it belongs to the foreground and edges are weighted for indicating the appearance compatibility of two neighboring nodes. This graph is dynamically learned and applied in object tracking and model updating. The proposed algorithm performs three main steps in each frame. First, the graph is initialized by assigning binary weights of some image patches to indicate the object and background patches according to the predicted bounding box. Second, the graph is optimized to refine the patch weights. Third, the object feature representation is updated by imposing the weights of patches on the extracted image features. The object location is predicted by maximizing the classification score in the structured support vector machine. Extensive experiments show that the proposed tracking algorithm performs well against the state-of-the-art methods.",
author = "Chenglong Li and Liang Lin and Wangmeng Zuo and Jin Tang and Yang, {Ming Hsuan}",
year = "2018",
month = "8",
day = "10",
doi = "10.1109/TPAMI.2018.2864965",
language = "English",
journal = "IEEE Transactions on Pattern Analysis and Machine Intelligence",
issn = "0162-8828",
publisher = "IEEE Computer Society",

}

Visual Tracking via Dynamic Graph Learning. / Li, Chenglong; Lin, Liang; Zuo, Wangmeng; Tang, Jin; Yang, Ming Hsuan.

In: IEEE transactions on pattern analysis and machine intelligence, 10.08.2018.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Visual Tracking via Dynamic Graph Learning

AU - Li, Chenglong

AU - Lin, Liang

AU - Zuo, Wangmeng

AU - Tang, Jin

AU - Yang, Ming Hsuan

PY - 2018/8/10

Y1 - 2018/8/10

N2 - Existing visual tracking methods usually localize a target object with a bounding box, in which the performance of the foreground object trackers or detectors is often affected by the inclusion of background clutter. To handle this problem, we learn a patch-based graph representation for visual tracking. The tracked object is modeled by with a graph by taking a set of non-overlapping image patches as nodes, in which the weight of each node indicates how likely it belongs to the foreground and edges are weighted for indicating the appearance compatibility of two neighboring nodes. This graph is dynamically learned and applied in object tracking and model updating. The proposed algorithm performs three main steps in each frame. First, the graph is initialized by assigning binary weights of some image patches to indicate the object and background patches according to the predicted bounding box. Second, the graph is optimized to refine the patch weights. Third, the object feature representation is updated by imposing the weights of patches on the extracted image features. The object location is predicted by maximizing the classification score in the structured support vector machine. Extensive experiments show that the proposed tracking algorithm performs well against the state-of-the-art methods.

AB - Existing visual tracking methods usually localize a target object with a bounding box, in which the performance of the foreground object trackers or detectors is often affected by the inclusion of background clutter. To handle this problem, we learn a patch-based graph representation for visual tracking. The tracked object is modeled by with a graph by taking a set of non-overlapping image patches as nodes, in which the weight of each node indicates how likely it belongs to the foreground and edges are weighted for indicating the appearance compatibility of two neighboring nodes. This graph is dynamically learned and applied in object tracking and model updating. The proposed algorithm performs three main steps in each frame. First, the graph is initialized by assigning binary weights of some image patches to indicate the object and background patches according to the predicted bounding box. Second, the graph is optimized to refine the patch weights. Third, the object feature representation is updated by imposing the weights of patches on the extracted image features. The object location is predicted by maximizing the classification score in the structured support vector machine. Extensive experiments show that the proposed tracking algorithm performs well against the state-of-the-art methods.

UR - http://www.scopus.com/inward/record.url?scp=85052586220&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85052586220&partnerID=8YFLogxK

U2 - 10.1109/TPAMI.2018.2864965

DO - 10.1109/TPAMI.2018.2864965

M3 - Article

C2 - 30106707

AN - SCOPUS:85052586220

JO - IEEE Transactions on Pattern Analysis and Machine Intelligence

JF - IEEE Transactions on Pattern Analysis and Machine Intelligence

SN - 0162-8828

ER -