Learning Fine-Grained Similarity Matching Networks for Visual Tracking

Zhang, Dawei; Zheng, Zhonglong; He, Xiaowei; Su, Liu; Chen, Liyuan

doi:10.1145/3372278.3390729

Cited by 5 publications

(3 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In terms of trackers based on the Siamese network, Q. used a residual attention network to capture more robust object features. Shen et al (2019) develop a hierarchical attention Siamese network for visual tracking, while Zhang et al (2020) utilize a series of structures to emphasize important semantic information. Tan et al (2021) develops a target-aware non-local block to leverage the long-range dependency, and a location-aware non-local block to associate multiple response maps.…”

Section: The Tracking Algorithm Based On the Attention Mechanismmentioning

confidence: 99%

Object Tracking Based on Global Context Attention

Wang

Chen

Mao

et al. 2021

International Journal of Cognitive Informatics and Natural Intelligence

View full text Add to dashboard Cite

Previous research has shown that tracking algorithms cannot capture long-distance information and lead to the loss of the object when the object was deformed, the illumination changed, and the background was disturbed by similar objects. To remedy this, this article proposes an object-tracking method by introducing the Global Context attention module into the Multi-Domain Network (MDNet) tracker. This method can learn the robust feature representation of the object through the Global Context attention module to better distinguish the background from the object in the presence of interference factors. Extensive experiments on OTB2013, OTB2015, and UAV20L datasets show that the proposed method is significantly improved compared with MDNet and has competitive performance compared with more mainstream tracking algorithms. At the same time, the method proposed in this article achieves better results when the video sequence contains object deformation, illumination change, and background interference with similar objects.

show abstract

Section: The Tracking Algorithm Based On the Attention Mechanismmentioning

confidence: 99%

Object Tracking Based on Global Context Attention

Wang

Chen

Mao

et al. 2021

International Journal of Cognitive Informatics and Natural Intelligence

View full text Add to dashboard Cite

show abstract

“…Both can perform far beyond real-time online tracking without extra fine-tuning. Owing to the characteristics (neat, simple and efficient) in SiamFC, there are numerous follow-up improvements [ 13 , 35 , 36 , 37 ]. RASNet [ 13 ] explored a residual attention Siamese network to adapt the offline learned features representation to the tracked target, while SiamRPN [ 14 ] introduced a region proposal network into Siamese networks to simultaneously perform classification and regression for high-performance tracking.…”

Section: Related Workmentioning

confidence: 99%

HROM: Learning High-Resolution Representation and Object-Aware Masks for Visual Object Tracking

Zhang

Zheng

Wang

et al. 2020

Sensors

Self Cite

View full text Add to dashboard Cite

Siamese network-based trackers consider tracking as features cross-correlation between the target template and the search region. Therefore, feature representation plays an important role for constructing a high-performance tracker. However, all existing Siamese networks extract the deep but low-resolution features of the entire patch, which is not robust enough to estimate the target bounding box accurately. In this work, to address this issue, we propose a novel high-resolution Siamese network, which connects the high-to-low resolution convolution streams in parallel as well as repeatedly exchanges the information across resolutions to maintain high-resolution representations. The resulting representation is semantically richer and spatially more precise by a simple yet effective multi-scale feature fusion strategy. Moreover, we exploit attention mechanisms to learn object-aware masks for adaptive feature refinement, and use deformable convolution to handle complex geometric transformations. This makes the target more discriminative against distractors and background. Without bells and whistles, extensive experiments on popular tracking benchmarks containing OTB100, UAV123, VOT2018 and LaSOT demonstrate that the proposed tracker achieves state-of-the-art performance and runs in real time, confirming its efficiency and effectiveness.

show abstract

“…Recently, Siamese networks based trackers [1,16,25,26,44,45,51,52] have drawn much attention in tracking community. For this category of trackers, It formulates the visual object tracking problem as learning a general response map of similarity scores by cross-correlation between the feature representations learned for the target template and the search region.…”

Section: Introductionmentioning

confidence: 99%

Reinforced Similarity Learning

Zhang

Zheng

et al. 2020

Proceedings of the 28th ACM International Conference on Multimedia

Self Cite

View full text Add to dashboard Cite

Recently, Siamese networks based tracking algorithms have shown favorable performance. Latest work focuses on better feature embedding and target state estimation, which greatly improves the accuracy. Nevertheless, the simple cross-correlation operation of the features between a fixed template and the search region limits their robustness and discrimination capability. In this paper, we pay more attention to learn an outstanding similarity measure for robust tracking. We propose a novel relation network that can be integrated on top of previous trackers without any need for further training of the siamese networks, which achieves a superior discriminative ability. During online inference, we utilize the feedback from high-confidence tracking results to obtain an additional template and update it, which improves the robustness and generalization. We implement two versions of the proposed approach with the SiamFC-based tracker and SiamRPN-based tracker to validate the strong compatibility of our algorithm. Extensive experimental results on several tracking benchmarks indicate that the proposed method can effectively improve the performance and robustness of the underlying trackers without reducing speed too much, and performs superiorly against the state-of-the-art trackers.

show abstract

Learning Fine-Grained Similarity Matching Networks for Visual Tracking

Cited by 5 publications

References 10 publications

Object Tracking Based on Global Context Attention

Object Tracking Based on Global Context Attention

HROM: Learning High-Resolution Representation and Object-Aware Masks for Visual Object Tracking

Reinforced Similarity Learning

Contact Info

Product

Resources

About