Learning Aberrance Repressed Correlation Filters for Real-Time UAV Tracking

Huang, Ziyuan; Fu, Changhong; Li, Yiming; Lin, Fuling; Lu, Peng

doi:10.1109/iccv.2019.00298

Cited by 332 publications

(250 citation statements)

References 24 publications

Supporting

Mentioning

215

Contrasting

Order By: Relevance

“…In this section, the proposed SARCT framework is systematically evaluated on three UAV tracking benchmarks: 1) UAVDT [1]; 2) UAV123 [9]; and 3) DTB [23], which totally include 243 challenging image sequences altogether involving over 90 000 frames. Following the protocol used in recently published methods [2], [7], [22], the results in one-pass evaluation (OPE) [36] are reported. The evaluation is based on two performance metrics: 1) success plot and 2) precision plot.…”

Section: Experimental Evaluationmentioning

confidence: 99%

“…As a distinct approach, the implementation ECO_SARCT of the SARCT framework is taken for this comparative study, against state-ofthe-art algorithms on the DTB dataset. The compared tracking methods are ARCF [2], TADT [37], LDES [38], MCCT [39], STRCF [40], SiamFC-tri [41], CSRDCF [42], and BACF [43]. Among them, BACF, ARCF, CSRDCF, LDES, MCCT, and STRCF are DCF trackers, while SiamFC-tri and TADT are based on the end-to-end CNN.…”

Section: Comparison With State-of-the-art Trackers 1) Experimentalmentioning

confidence: 99%

“…S IGNIFICANT developments in unmanned aerial vehicles (UAVs) have been witnessed recently, delivering more diversity and flexibility for photography than common surveillance cameras with fixed camera angles, scale, and view [1]. Aerial object tracking [2] has enabled many new and important applications in computer vision [3], such as crowd monitoring, target following, and aerial navigation. Aiming to analyze the movement of a certain target, visual tracking algorithms [4]- [6] locate their bounding boxes on the video stream according to the given initial state in the first frame.…”

Section: Introductionmentioning

confidence: 99%

“…Aiming to analyze the movement of a certain target, visual tracking algorithms [4]- [6] locate their bounding boxes on the video stream according to the given initial state in the first frame. Although methods based on the correlation filter [4] and Siamese network [5], [6] have achieved decent performance in general scenes, robust and accurate aerial tracking [2], [7] remains active research because of the particular challenging factors, such as unpredictable weather conditions, changing flying altitude, and shaking camera views. To be more specific, aerial objects are usually tiny and move very fast or rotate drastically, resulting in difficulties in tracking them.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Semantic-Aware Real-Time Correlation Tracking Framework for UAV Videos

Xue

Yin

et al. 2022

IEEE Trans. Cybern.

View full text Add to dashboard Cite

Discriminative correlation filter (DCF) has contributed tremendously to address the problem of object tracking benefitting from its high computational efficiency. However, it has suffered from performance degradation in unmanned aerial vehicle (UAV) tracking. This article presents a novel semanticaware real-time correlation tracking framework (SARCT) for UAV videos to enhance the performance of DCF trackers without incurring excessive computing cost. Specifically, SARCT first constructs an additional detection module to generate ROI proposals and to filter any response regarding the target irrelevant area. Then, a novel semantic segmentation module based on semantic template generation and semantic coefficient prediction is further introduced to capture semantic information, which can provide precise ROI mask, thereby effectively suppressing background interference in the ROI proposals. By sharing features and specific network layers for object detection and semantic segmentation, SARCT reduces parameter redundancy to attain sufficient speed for real-time applications. Systematic experiments are conducted on three typical aerial datasets in order to evaluate the performance of the proposed SARCT. The results demonstrate that SARCT is able to improve the accuracy of conventional DCF-based trackers significantly, outperforming state-of-the-art deep trackers.

show abstract

Section: Experimental Evaluationmentioning

confidence: 99%

Section: Comparison With State-of-the-art Trackers 1) Experimentalmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Semantic-Aware Real-Time Correlation Tracking Framework for UAV Videos

Xue

Yin

et al. 2022

IEEE Trans. Cybern.

View full text Add to dashboard Cite

show abstract

“…Gradually, deep learning has become a mainstream feature in tracking methods. Consequently, the latest correlation filter methods [ 30 , 31 ] make use of depth features due to their powerful representation capabilities.…”

Section: Related Workmentioning

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

Robust visual tracking via adaptive feature channel selection

Zhang

Hou

et al. 2022

Int J of Intelligent Sys

View full text Add to dashboard Cite

Discriminative correlation filters (DCFs) have shown promising tracking performance in recent years thanks to the powerful representation ability of deep features.However, a large number of target-irrelevant channels in deep features limits the tracking performance and increases the computational cost. To eliminate the negative impact of noisy channels and improve the utilization efficiency of deep features in DCF-based trackers, we present an adaptive feature channel selection method for robust visual tracking. Our method adaptively chooses the most discriminative channels to learn a more robust target appearance model, which is achieved by evaluating the energy relationship between background and foreground in each feature channel. Moreover, according to the feedback of channel selection, an adaptive model update strategy is proposed to alleviate the model degradation problem caused by incorrect model updating. Extensive experimental results obtained on five popular tracking benchmarks demonstrate the effectiveness of the proposed algorithm and its superiority over the state-of-the-art trackers.

show abstract

Learning Aberrance Repressed Correlation Filters for Real-Time UAV Tracking

Cited by 332 publications

References 24 publications

Semantic-Aware Real-Time Correlation Tracking Framework for UAV Videos

Semantic-Aware Real-Time Correlation Tracking Framework for UAV Videos

A Visual Tracker Offering More Solutions

Robust visual tracking via adaptive feature channel selection

Contact Info

Product

Resources

About