Abstract
Object tracking belongs to active research areas in computer vision. We are interested in matching-based trackers exploiting deep machine learning known as Siamese trackers. Their powerful capabilities stem from similarity learning. This tracking paradigm is promising due to its inherent balance between performance and efficiency, so trackers of this type are suitable for real-time generic object tracking. There is an upsurge in research interest in Siamese trackers and the lack of available specialized surveys in this category. In this survey, we aim to identify and elaborate on the most significant challenges the Siamese trackers face. Our goal is to answer what design decisions the authors made and what problems they attempted to solve in the first place. We thus perform an in-depth analysis of the core principles on which Siamese trackers operate with a discussion of incentives behind them. Besides, we provide an up-to-date qualitative and quantitative comparison of the prominent Siamese trackers on established benchmarks. Among other things, we discuss current trends in developing Siamese trackers. Our survey could help absorb the details about the underlying principles of Siamese trackers and the challenges they face.
Highlights
O BJECT tracking is among very active research areas in the field of computer vision [1]
The logistic loss was adopted in trackers Siamese fully convolutional network (SiamFC) [12], Correlation filter network (CFNet) [54] and Dynamic Siamese network (DSiam) [55]. This loss was a foundation to many upcoming trackers, e.g., Siamese classification and regression networks (SiamCAR) [20] or Foreground information guidance for Siamese visual tracking (FIGSiam) [56], where they extended this loss by adding more terms
A very similar architecture to SiamCAR that we introduced in section IV-A is Fully convolutional anchor-free Siamese network (FCAF) [67]
Summary
O BJECT tracking is among very active research areas in the field of computer vision [1]. Models based on deep learning are strong at distinguishing objects of different categories with good generalization capabilities [13] With this in mind, the key is to find features that simultaneously allow differentiating between an object and a background and allow handling changes of the tracked object, even when not known a priori [16]. We strive for deep analysis of a given tracking paradigm rather than to provide a comprehensive, broad discussion covering the vast population of approaches to object tracking For this purpose, other works complement our contribution (section II). The main purpose of this survey is to convey the main properties of Siamese trackers It provides a discussion about object tracking in general.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
