‘Skimming-Perusal’ Tracking: A Framework for Real-Time and Robust Long-Term Tracking

Yan, Bin; Zhao, Haojie; Wang, Dong; Lu, Huchuan; Yang, Xiaoyun

doi:10.1109/iccv.2019.00247

Cited by 193 publications

(114 citation statements)

References 40 publications

Supporting

Mentioning

114

Contrasting

Order By: Relevance

“…The target then reappears in a random location, which requires the tracker to have a global search capability to recapture the target quickly. Figure 13 compares our tracker with the recently proposed advanced trackers DiMP and SPLT [ 64 ], as well as DaSiamRPN [ 41 ], using a video sequence from the VOT2018-LT dataset. DiMP, with the same online learning ability as our tracker, exhibits a significantly higher accuracy than that of SPLT and DaSiamRPN using offline learning.…”

Section: Experiments and Discussionmentioning

confidence: 99%

A Visual Tracker Offering More Solutions

Zhao

Mahmoud

Ren

et al. 2020

Sensors

View full text Add to dashboard Cite

Most trackers focus solely on robustness and accuracy. Visual tracking, however, is a long-term problem with a high time limitation. A tracker that is robust, accurate, with long-term sustainability and real-time processing, is of high research value and practical significance. In this paper, we comprehensively consider these requirements in order to propose a new, state-of-the-art tracker with an excellent performance. EfficientNet-B0 is adopted for the first time via neural architecture search technology as the backbone network for the tracking task. This improves the network feature extraction ability and significantly reduces the number of parameters required for the tracker backbone network. In addition, maximal Distance Intersection-over-Union is set as the target estimation method, enhancing network stability and increasing the offline training convergence rate. Channel and spatial dual attention mechanisms are employed in the target classification module to improve the discrimination of the trackers. Furthermore, the conjugate gradient optimization strategy increases the speed of the online learning target classification module. A two-stage search method combined with a screening module is proposed to enable the tracker to cope with sudden target movement and reappearance following a brief disappearance. Our proposed method has an obvious speed advantage compared with pure global searching and achieves an optimal performance on OTB2015, VOT2016, VOT2018-LT, UAV-123 and LaSOT while running at over 50 FPS.

show abstract

Section: Experiments and Discussionmentioning

confidence: 99%

A Visual Tracker Offering More Solutions

Zhao

Mahmoud

Ren

et al. 2020

Sensors

View full text Add to dashboard Cite

show abstract

“…Five different sizes of the bounding box are selected from the results of dimension clustering on the one scale. As for this paper, we need to complete K-means clustering [29] for VLD-45-S on the different scales. Besides, our method makes sure the number of bounding box on a scale.…”

Section: Improvement Based On the Algorithm Of Single-stagementioning

confidence: 99%

Single Stage Vehicle Logo Detector Based on Multi-Scale Prediction

Zhang

Yang

et al. 2020

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

Vehicle logo detection technology is one of the research directions in the application of intelligent transportation systems. It is an important extension of detection technology based on license plates and motorcycle types. A vehicle logo is characterized by uniqueness, conspicuousness, and diversity. Therefore, thorough research is important in theory and application. Although there are some related works for object detection, most of them cannot achieve real-time detection for different scenes. Meanwhile, some real-time detection methods of single-stage have performed poorly in the object detection of small sizes. In order to solve the problem that the training samples are scarce, our work in this paper is improved by constructing the data of a vehicle logo (VLD-45-S), multistage pre-training, multi-scale prediction, feature fusion between deeper with shallow layer, dimension clustering of the bounding box, and multiscale detection training. On the basis of keeping speed, this article improves the detection precision of the vehicle logo. The generalization of the detection model and anti-interference capability in real scenes are optimized by data enrichment. Experimental results show that the accuracy and speed of the detection algorithm are improved for the object of small sizes.

show abstract

“…However, the original RNN [51] has encountered vanished gradient problem when the image sequence is too long, which may hinder information learning spanning over a long sequence. Yan et al [52] developed an algorithm for robust long-term tracking, called ''Skimming-Perusal'' Tracking (SPLT), which can effectively choose the most possible regions from a large number of sliding windows. Simultaneously, the tracking performance is largely dependent on the object detection, so the bottleneck of detection limits the improvement of object tracking.…”

Section: ) Deep Learning Trackersmentioning

confidence: 99%

ADT: Object Tracking Algorithm Based on Adaptive Detection

Ming

Zhang

2020

IEEE Access

View full text Add to dashboard Cite

Object tracking is one of the most fundamental and important fields in computer vision with a wide range of applications. Although great progress has been made in object tracking combined with detection, there is still enormous challenges in real-time applications and for the computer cannot effectively capture the temporal correlations of targets and background clutter. In order to improve the performance of tracking algorithms under complex unconstrained conditions, we propose a novel tracking framework based on adaptive detection, called adaptive detection tracking (ADT). First, we exploit the temporal correlation of the recurrent neural network to predict the target's motion direction and efficiently update the region of interest (RoI) in the narrow range of the next frame. Then, the algorithm utilizes the correlation filter to initialize the defined region of interest based on the threshold. If the Interaction of Union (IoU) of the predicted bounding box and the groundtruth bounding box is greater than the set threshold, the predicted bounding box will be directly output as the tracking results, whereas the detection is adaptively carried out in the determined RoI. Finally, the predicted bounding box refines the direction model as the input of the next frame to complete the whole tracking flow. Our proposed adaptive detection tracking mechanism can efficiently realize non-frame-by-frame adaptive detection with excellent tracking accuracy and is more robust in the unconstrained scenes, especially for occlusion. Comprehensive experiments demonstrate that our approach consistently achieves state-of-the-art results and runs in real-time on six large tracking benchmarks, including OTB100, VOT2016, VOT2017, TC128, UAV123 and LaSOT datasets. INDEX TERMS Recurrent neural network, adaptive detection, object tracking, correlation filtering, model compression. YUE MING received the B.S. degree in communication engineering, the M.Sc. degree in humancomputer interaction engineering, and the Ph.D. degree in signal and information processing from

show abstract

‘Skimming-Perusal’ Tracking: A Framework for Real-Time and Robust Long-Term Tracking

Cited by 193 publications

References 40 publications

A Visual Tracker Offering More Solutions

A Visual Tracker Offering More Solutions

Single Stage Vehicle Logo Detector Based on Multi-Scale Prediction

ADT: Object Tracking Algorithm Based on Adaptive Detection

Contact Info

Product

Resources

About