NUS-PRO: A New Visual Tracking Challenge

Annan Li, Min Lin, Yi Wu, Ming Hsuan Yang, Shuicheng Yan

Research output: Contribution to journalArticle

66 Citations (Scopus)

Abstract

Numerous approaches on object tracking have been proposed during the past decade with demonstrated success. However, most tracking algorithms are evaluated on limited video sequences and annotations. For thorough performance evaluation, we propose a large-scale database which contains 365 challenging image sequences of pedestrians and rigid objects. The database covers 12 kinds of objects, and most of the sequences are captured from moving cameras. Each sequence is annotated with target location and occlusion level for evaluation. A thorough experimental evaluation of 20 state-of-the-art tracking algorithms is presented with detailed analysis using different metrics. The database is publicly available and evaluation can be carried out online for fair assessments of visual tracking algorithms.

Original languageEnglish
Article number7072555
Pages (from-to)335-349
Number of pages15
JournalIEEE transactions on pattern analysis and machine intelligence
Volume38
Issue number2
DOIs
Publication statusPublished - 2016 Feb 1

Fingerprint

Visual Tracking
Object Tracking
Evaluation
Image Sequence
Occlusion
Experimental Evaluation
Annotation
Performance Evaluation
Camera
Cameras
Cover
Metric
Target
Object

All Science Journal Classification (ASJC) codes

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

Cite this

Li, Annan ; Lin, Min ; Wu, Yi ; Yang, Ming Hsuan ; Yan, Shuicheng. / NUS-PRO : A New Visual Tracking Challenge. In: IEEE transactions on pattern analysis and machine intelligence. 2016 ; Vol. 38, No. 2. pp. 335-349.
@article{556256eb23664a2c8b5fff51bc2f7608,
title = "NUS-PRO: A New Visual Tracking Challenge",
abstract = "Numerous approaches on object tracking have been proposed during the past decade with demonstrated success. However, most tracking algorithms are evaluated on limited video sequences and annotations. For thorough performance evaluation, we propose a large-scale database which contains 365 challenging image sequences of pedestrians and rigid objects. The database covers 12 kinds of objects, and most of the sequences are captured from moving cameras. Each sequence is annotated with target location and occlusion level for evaluation. A thorough experimental evaluation of 20 state-of-the-art tracking algorithms is presented with detailed analysis using different metrics. The database is publicly available and evaluation can be carried out online for fair assessments of visual tracking algorithms.",
author = "Annan Li and Min Lin and Yi Wu and Yang, {Ming Hsuan} and Shuicheng Yan",
year = "2016",
month = "2",
day = "1",
doi = "10.1109/TPAMI.2015.2417577",
language = "English",
volume = "38",
pages = "335--349",
journal = "IEEE Transactions on Pattern Analysis and Machine Intelligence",
issn = "0162-8828",
publisher = "IEEE Computer Society",
number = "2",

}

NUS-PRO : A New Visual Tracking Challenge. / Li, Annan; Lin, Min; Wu, Yi; Yang, Ming Hsuan; Yan, Shuicheng.

In: IEEE transactions on pattern analysis and machine intelligence, Vol. 38, No. 2, 7072555, 01.02.2016, p. 335-349.

Research output: Contribution to journalArticle

TY - JOUR

T1 - NUS-PRO

T2 - A New Visual Tracking Challenge

AU - Li, Annan

AU - Lin, Min

AU - Wu, Yi

AU - Yang, Ming Hsuan

AU - Yan, Shuicheng

PY - 2016/2/1

Y1 - 2016/2/1

N2 - Numerous approaches on object tracking have been proposed during the past decade with demonstrated success. However, most tracking algorithms are evaluated on limited video sequences and annotations. For thorough performance evaluation, we propose a large-scale database which contains 365 challenging image sequences of pedestrians and rigid objects. The database covers 12 kinds of objects, and most of the sequences are captured from moving cameras. Each sequence is annotated with target location and occlusion level for evaluation. A thorough experimental evaluation of 20 state-of-the-art tracking algorithms is presented with detailed analysis using different metrics. The database is publicly available and evaluation can be carried out online for fair assessments of visual tracking algorithms.

AB - Numerous approaches on object tracking have been proposed during the past decade with demonstrated success. However, most tracking algorithms are evaluated on limited video sequences and annotations. For thorough performance evaluation, we propose a large-scale database which contains 365 challenging image sequences of pedestrians and rigid objects. The database covers 12 kinds of objects, and most of the sequences are captured from moving cameras. Each sequence is annotated with target location and occlusion level for evaluation. A thorough experimental evaluation of 20 state-of-the-art tracking algorithms is presented with detailed analysis using different metrics. The database is publicly available and evaluation can be carried out online for fair assessments of visual tracking algorithms.

UR - http://www.scopus.com/inward/record.url?scp=84962116001&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84962116001&partnerID=8YFLogxK

U2 - 10.1109/TPAMI.2015.2417577

DO - 10.1109/TPAMI.2015.2417577

M3 - Article

AN - SCOPUS:84962116001

VL - 38

SP - 335

EP - 349

JO - IEEE Transactions on Pattern Analysis and Machine Intelligence

JF - IEEE Transactions on Pattern Analysis and Machine Intelligence

SN - 0162-8828

IS - 2

M1 - 7072555

ER -