Real-time object tracking via online discriminative feature selection

Kaihua Zhang, Lei Zhang, Ming Hsuan Yang

Research output: Contribution to journalArticle

143 Citations (Scopus)

Abstract

Most tracking-by-detection algorithms train discriminative classifiers to separate target objects from their surrounding background. In this setting, noisy samples are likely to be included when they are not properly sampled, thereby causing visual drift. The multiple instance learning (MIL) paradigm has been recently applied to alleviate this problem. However, important prior information of instance labels and the most correct positive instance (i.e., the tracking result in the current frame) can be exploited using a novel formulation much simpler than an MIL approach. In this paper, we show that integrating such prior information into a supervised learning algorithm can handle visual drift more effectively and efficiently than the existing MIL tracker. We present an online discriminative feature selection algorithm that optimizes the objective function in the steepest ascent direction with respect to the positive samples while in the steepest descent direction with respect to the negative ones. Therefore, the trained classifier directly couples its score with the importance of samples, leading to a more robust and efficient tracker. Numerous experimental evaluations with state-of-the-art algorithms on challenging sequences demonstrate the merits of the proposed algorithm.

Original languageEnglish
Article number6576884
Pages (from-to)4664-4677
Number of pages14
JournalIEEE Transactions on Image Processing
Volume22
Issue number12
DOIs
Publication statusPublished - 2013 Oct 7

Fingerprint

Feature extraction
Classifiers
Supervised learning
Learning algorithms
Labels

All Science Journal Classification (ASJC) codes

  • Software
  • Computer Graphics and Computer-Aided Design

Cite this

@article{b5e6f87a34e44ed492e938850decf9bf,
title = "Real-time object tracking via online discriminative feature selection",
abstract = "Most tracking-by-detection algorithms train discriminative classifiers to separate target objects from their surrounding background. In this setting, noisy samples are likely to be included when they are not properly sampled, thereby causing visual drift. The multiple instance learning (MIL) paradigm has been recently applied to alleviate this problem. However, important prior information of instance labels and the most correct positive instance (i.e., the tracking result in the current frame) can be exploited using a novel formulation much simpler than an MIL approach. In this paper, we show that integrating such prior information into a supervised learning algorithm can handle visual drift more effectively and efficiently than the existing MIL tracker. We present an online discriminative feature selection algorithm that optimizes the objective function in the steepest ascent direction with respect to the positive samples while in the steepest descent direction with respect to the negative ones. Therefore, the trained classifier directly couples its score with the importance of samples, leading to a more robust and efficient tracker. Numerous experimental evaluations with state-of-the-art algorithms on challenging sequences demonstrate the merits of the proposed algorithm.",
author = "Kaihua Zhang and Lei Zhang and Yang, {Ming Hsuan}",
year = "2013",
month = "10",
day = "7",
doi = "10.1109/TIP.2013.2277800",
language = "English",
volume = "22",
pages = "4664--4677",
journal = "IEEE Transactions on Image Processing",
issn = "1057-7149",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "12",

}

Real-time object tracking via online discriminative feature selection. / Zhang, Kaihua; Zhang, Lei; Yang, Ming Hsuan.

In: IEEE Transactions on Image Processing, Vol. 22, No. 12, 6576884, 07.10.2013, p. 4664-4677.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Real-time object tracking via online discriminative feature selection

AU - Zhang, Kaihua

AU - Zhang, Lei

AU - Yang, Ming Hsuan

PY - 2013/10/7

Y1 - 2013/10/7

N2 - Most tracking-by-detection algorithms train discriminative classifiers to separate target objects from their surrounding background. In this setting, noisy samples are likely to be included when they are not properly sampled, thereby causing visual drift. The multiple instance learning (MIL) paradigm has been recently applied to alleviate this problem. However, important prior information of instance labels and the most correct positive instance (i.e., the tracking result in the current frame) can be exploited using a novel formulation much simpler than an MIL approach. In this paper, we show that integrating such prior information into a supervised learning algorithm can handle visual drift more effectively and efficiently than the existing MIL tracker. We present an online discriminative feature selection algorithm that optimizes the objective function in the steepest ascent direction with respect to the positive samples while in the steepest descent direction with respect to the negative ones. Therefore, the trained classifier directly couples its score with the importance of samples, leading to a more robust and efficient tracker. Numerous experimental evaluations with state-of-the-art algorithms on challenging sequences demonstrate the merits of the proposed algorithm.

AB - Most tracking-by-detection algorithms train discriminative classifiers to separate target objects from their surrounding background. In this setting, noisy samples are likely to be included when they are not properly sampled, thereby causing visual drift. The multiple instance learning (MIL) paradigm has been recently applied to alleviate this problem. However, important prior information of instance labels and the most correct positive instance (i.e., the tracking result in the current frame) can be exploited using a novel formulation much simpler than an MIL approach. In this paper, we show that integrating such prior information into a supervised learning algorithm can handle visual drift more effectively and efficiently than the existing MIL tracker. We present an online discriminative feature selection algorithm that optimizes the objective function in the steepest ascent direction with respect to the positive samples while in the steepest descent direction with respect to the negative ones. Therefore, the trained classifier directly couples its score with the importance of samples, leading to a more robust and efficient tracker. Numerous experimental evaluations with state-of-the-art algorithms on challenging sequences demonstrate the merits of the proposed algorithm.

UR - http://www.scopus.com/inward/record.url?scp=84884832249&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84884832249&partnerID=8YFLogxK

U2 - 10.1109/TIP.2013.2277800

DO - 10.1109/TIP.2013.2277800

M3 - Article

C2 - 23955750

AN - SCOPUS:84884832249

VL - 22

SP - 4664

EP - 4677

JO - IEEE Transactions on Image Processing

JF - IEEE Transactions on Image Processing

SN - 1057-7149

IS - 12

M1 - 6576884

ER -