Spatial and semantic convolutional features for robust visual object tracking

• Published in: Multimedia tools and applications Vol. 79; no. 21-22; pp. 15095 - 15115
• Main Authors: Zhang, Jianming, Jin, Xiaokang, Sun, Juan, Wang, Jin, Sangaiah, Arun Kumar
• Format: Journal Article
• Language: English
• Published: New York Springer US 01.06.2020; Springer Nature B.V
• Subjects: Artificial neural networks; Computer Communication Networks; Computer Science; Computer vision; Data Structures and Information Theory; Feature extraction; Model updating; Multimedia Information Systems; Multiscale analysis; Optical tracking; Semantics; Sidelobes; Special Purpose and Application-Based Systems; Variation; Correlation filter; Scale adaptive; Object tracking; Convolutional neural networks; Model updating strategy
• ISSN: 1380-7501; 1573-7721
• DOI: 10.1007/s11042-018-6562-8

Robust and accurate visual tracking is a challenging problem in computer vision. In this paper, we exploit spatial and semantic convolutional features extracted from convolutional neural networks in continuous object tracking. The spatial features retain higher resolution for precise localization and semantic features capture more semantic information and less fine-grained spatial details. Therefore, we localize the target by fusing these different features, which improves the tracking accuracy. Besides, we construct the multi-scale pyramid correlation filter of the target and extract its spatial features. This filter determines the scale level effectively and tackles target scale estimation. Finally, we further present a novel model updating strategy, and exploit peak sidelobe ratio (PSR) and skewness to measure the comprehensive fluctuation of response map for efficient tracking performance. Each contribution above is validated on 50 image sequences of tracking benchmark OTB-2013. The experimental comparison shows that our algorithm performs favorably against 12 state-of-the-art trackers.