Share Email Print

Optical Engineering • Open Access • new

Robust visual tracking via multiscale deep sparse networks
Author(s): Xin Wang; Zhiqiang Hou; Wangsheng Yu; Yang Xue; Zefenfen Jin; Bo Dai

Paper Abstract

In visual tracking, deep learning with offline pretraining can extract more intrinsic and robust features. It has significant success solving the tracking drift in a complicated environment. However, offline pretraining requires numerous auxiliary training datasets and is considerably time-consuming for tracking tasks. To solve these problems, a multiscale sparse networks-based tracker (MSNT) under the particle filter framework is proposed. Based on the stacked sparse autoencoders and rectifier linear unit, the tracker has a flexible and adjustable architecture without the offline pretraining process and exploits the robust and powerful features effectively only through online training of limited labeled data. Meanwhile, the tracker builds four deep sparse networks of different scales, according to the target’s profile type. During tracking, the tracker selects the matched tracking network adaptively in accordance with the initial target’s profile type. It preserves the inherent structural information more efficiently than the single-scale networks. Additionally, a corresponding update strategy is proposed to improve the robustness of the tracker. Extensive experimental results on a large scale benchmark dataset show that the proposed method performs favorably against state-of-the-art methods in challenging environments.

Paper Details

Date Published: 21 April 2017
PDF: 14 pages
Opt. Eng. 56(4) 043107 doi: 10.1117/1.OE.56.4.043107
Published in: Optical Engineering Volume 56, Issue 4
Show Author Affiliations
Xin Wang, Air Force Engineering Univ. (China)
Zhiqiang Hou, Air Force Engineering Univ. (China)
Wangsheng Yu, Air Force Engineering Univ. (China)
Yang Xue, Air Force Engineering Univ. (China)
Zefenfen Jin, Air Force Engineering Univ. (China)
Bo Dai, Air Force Engineering Univ. (China)

© SPIE. Terms of Use
Back to Top