Deep Learning for Visual Tracking: A Comprehensive Survey

Marvasti-Zadeh, Seyed Mojtaba; Cheng, Li; Ghanei-Yakhdan, Hossein; Kasaei, Shohreh

doi:10.1109/tits.2020.3046478

Cited by 235 publications

(126 citation statements)

References 228 publications

Supporting

Mentioning

102

Contrasting

Order By: Relevance

“…The most recent comprehensive survey is from 2021 by Marvasti-Zadeh et al [17]. This work broadly covers existing approaches to object tracking in general.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Siamese Visual Object Tracking: A Survey

Ondrašovič

Tarábek

2021

IEEE Access

View full text Add to dashboard Cite

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-todate 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.INDEX TERMS visual object tracking, deep learning, Siamese neural networks, similarity learning, fully convolutional networks

show abstract

“…The most recent comprehensive survey is from 2021 by Marvasti-Zadeh et al [17]. This work broadly covers existing approaches to object tracking in general.…”

Section: Related Workmentioning

confidence: 99%

“…Currently, popular visual tracking methods revolve around Siamese neural networks (section III). The Siamese-based networks are considered the most promising architectures based on their balance between performance and efficiency [17]. With this in mind, these architectures are the primary focus of this survey.…”

Section: Introductionmentioning

confidence: 99%

Siamese Visual Object Tracking: A Survey

Ondrašovič

Tarábek

2021

IEEE Access

View full text Add to dashboard Cite

show abstract

“…The visual tracking methods can be classified into generative tracking methods and discriminative tracking methods [31,40]. Among the discriminative-based trackers, the DCF promote the visual tracking to a new level.…”

Section: Related Workmentioning

confidence: 99%

“…Visual tracking aims to estimate the state of the target in image sequences, given its initial state. It plays a crucial role in computer vision-based applications, e.g., vehicle navigation, video surveillance and robotic perception [2,16,26,31]. In recent years, the DCF-based methods have attracted extensive attention due to the high efficiency.…”

Section: Introductionmentioning

confidence: 99%

Spatio-temporal joint aberrance suppressed correlation filter for visual tracking

Kim

Wang

et al. 2021

Complex Intell. Syst.

View full text Add to dashboard Cite

The discriminative correlation filter (DCF)-based tracking methods have achieved remarkable performance in visual tracking. However, the existing DCF paradigm still suffers from dilemmas such as boundary effect, filter degradation, and aberrance. To address these problems, we propose a spatio-temporal joint aberrance suppressed regularization (STAR) correlation filter tracker under a unified framework of response map. Specifically, a dynamic spatio-temporal regularizer is introduced into the DCF to alleviate the boundary effect and filter degradation, simultaneously. Meanwhile, an aberrance suppressed regularizer is exploited to reduce the interference of background clutter. The proposed STAR model is effectively optimized using the alternating direction method of multipliers (ADMM). Finally, comprehensive experiments on TC128, OTB2013, OTB2015 and UAV123 benchmarks demonstrate that the STAR tracker achieves compelling performance compared with the state-of-the-art (SOTA) trackers.

show abstract

“…Most previous VOT works using traditional approaches such as Kalman and particle filters [4,5], Discriminative Correlation Filter (DCF) [6], or silhouette tracking [7], simplify the tracking procedure by constraining the tracking scenarios with, for example, stationary cameras, limited number of objects, limited occlusions, or absence of sudden background or object appearance changes. These methods usually use handcrafted feature representations (e.g., Histogram of Gradients (HOG) [8], color, position) and their target modeling is not dynamic [9]. In real-world scenarios, however, such constraints are often not applicable and VOT methods based on these traditional approaches perform poorly.…”

Section: Introductionmentioning

confidence: 99%

Multiple Pedestrians and Vehicles Tracking in Aerial Imagery Using a Convolutional Neural Network

et al. 2021

View full text Add to dashboard Cite

In this paper, we address various challenges in multi-pedestrian and vehicle tracking in high-resolution aerial imagery by intensive evaluation of a number of traditional and Deep Learning based Single- and Multi-Object Tracking methods. We also describe our proposed Deep Learning based Multi-Object Tracking method AerialMPTNet that fuses appearance, temporal, and graphical information using a Siamese Neural Network, a Long Short-Term Memory, and a Graph Convolutional Neural Network module for more accurate and stable tracking. Moreover, we investigate the influence of the Squeeze-and-Excitation layers and Online Hard Example Mining on the performance of AerialMPTNet. To the best of our knowledge, we are the first to use these two for regression-based Multi-Object Tracking. Additionally, we studied and compared the L1 and Huber loss functions. In our experiments, we extensively evaluate AerialMPTNet on three aerial Multi-Object Tracking datasets, namely AerialMPT and KIT AIS pedestrian and vehicle datasets. Qualitative and quantitative results show that AerialMPTNet outperforms all previous methods for the pedestrian datasets and achieves competitive results for the vehicle dataset. In addition, Long Short-Term Memory and Graph Convolutional Neural Network modules enhance the tracking performance. Moreover, using Squeeze-and-Excitation and Online Hard Example Mining significantly helps for some cases while degrades the results for other cases. In addition, according to the results, L1 yields better results with respect to Huber loss for most of the scenarios. The presented results provide a deep insight into challenges and opportunities of the aerial Multi-Object Tracking domain, paving the way for future research.

show abstract

Deep Learning for Visual Tracking: A Comprehensive Survey

Cited by 235 publications

References 228 publications

Siamese Visual Object Tracking: A Survey

Siamese Visual Object Tracking: A Survey

Spatio-temporal joint aberrance suppressed correlation filter for visual tracking

Multiple Pedestrians and Vehicles Tracking in Aerial Imagery Using a Convolutional Neural Network

Contact Info

Product

Resources

About