One-Class SVM Assisted Accurate Tracking
Paper in proceeding, 2012

Recently, tracking is regarded as a binary classification problem by discriminative tracking methods. However, such binary classification may not fully handle the outliers, which may cause drifting. In this paper, we argue that tracking may be regarded as one-class problem, which avoids gathering limited negative samples for background description. Inspired by the fact the positive feature space generated by One-Class SVM is bounded by a closed sphere, we propose a novel tracking method utilizing One-Class SVMs that adopt HOG and 2 bit-BP as features, called One-Class SVM Tracker (OCST). Simultaneously an efficient initialization and online updating scheme is also proposed. Extensive experimental results prove that OCST outperforms some state-of-the-art discriminative tracking methods on providing accurate tracking and alleviating serious drifting.

visual tracking

image database retrieval

support vector machine

Author

Chen Gong

Yu Qiao

Jie Yang

Irene Yu-Hua Gu

Chalmers, Signals and Systems, Signal Processing and Biomedical Engineering

6th ACM/IEEE Int'l Conf on Distributed Smart Cameras (ICDSC 12), Oct 30 - Nov.2, 2012, Hong Kong

6 pages-

Areas of Advance

Transport

Life Science Engineering (2010-2018)

Subject Categories

Bioinformatics (Computational Biology)

Control Engineering

Signal Processing

Computer Vision and Robotics (Autonomous Systems)

Other Electrical Engineering, Electronic Engineering, Information Engineering

More information

Created

10/7/2017