LaSOT: A High-quality Large-scale Single Object Tracking Benchmark

Fan, Heng; Bai, Hexin; Lin, Liting; Yang, Fan; Chu, Peng; Deng, Ge; Yu, Sijia; Harshit, -; Huang, Mingzhen; Liu, Juehuan; Xu, Yong; Liao, Chunyuan; Lin, Yuan; Ling, Haibin

doi:10.1007/s11263-020-01387-y

Cited by 289 publications

(501 citation statements)

References 82 publications

Supporting

Mentioning

488

Contrasting

Unclassified

Order By: Relevance

“…Then, ablation study is presented to analyze the effects of each component in the proposed tracker. Furthermore, the experimental results of HROM are compared on several popular tracking benchmarks including OTB2015 [ 28 ], UAV123 [ 51 ], VOT2018 [ 52 ] and LaSOT [ 53 ] with the state-of-the-art tracking algorithms.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

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

Zhang

Zheng

Wang

et al. 2020

Sensors

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

Section: Methodsmentioning

confidence: 99%

“…LaSOT [ 53 ] is a recent released more challenging benchmark for object tracking, and provides a high-quality and large-scale dense annotations with a total of 1400 sequences and 280 sequences for testing. Specifically, it includes 70 categories and each one contains 20 videos.…”

Section: Methodsmentioning

confidence: 99%

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

Zhang

Zheng

Wang

et al. 2020

Sensors

View full text Add to dashboard Cite

show abstract

“…Then, we conduct ablation studies from two perspectives: 1) evaluate the impact of the size of long-term memory sample set on the performance of the model; 2) compare and analyze the important role of the update method of long-term memory sample set on the model. Finally, we evaluate our method on a large-scale longterm benchmark dataset LaSOT [36].…”

Section: Methodsmentioning

confidence: 99%

Long-Term Visual Object Tracking via Continual Learning

2019

View full text Add to dashboard Cite

Long-term visual tracking is one of the most challenging problems in computer vision and is closer to practical application needs. In long-term video sequences, tracking targets often undergo dramatic appearance changes over time due to various factors such as scale variation, illumination change, occlusions and so on. In this work, we propose a novel robust long-term tracking framework based on continual learning and dynamic sample set modules. We transform the online tracking process into a continual learning process of the target model, and continuously learn various appearance changes to adapt to different scenarios. The continual learning module distills the beneficial knowledge of the old network to the new network through warm-up and joint training to achieve a comprehensive and holistic memory of the target appearance. Combining the dynamic sample set can effectively balance the short-term memory and long-term memory of the model, and establish a near-complete target appearance description in the long-term dimension to cope with various challenging situations. Experimental results on the large-scale long-term benchmark datasets LaSOT and UAV20L show that the proposed method performs favourably against other state-of-the-art trackers. INDEX TERMS Long-term visual tracking, continual learning, warm-up and joint training, dynamic sample set.

show abstract

“…It was followed by the more recent Temple‐Colour [26] and DTB70 [27] datasets, while the former focus on colour videos and the latter collects the videos from UAV or drone. Recently, larger tracking datasets have also been proposed [28, 29]. The VOT [30, 31] challenge updates its dataset yearly, with its own evaluation metrics.…”

Section: Related Workmentioning

confidence: 99%

End‐to‐end learning interpolation for object tracking in low frame‐rate video

Liu

Cao

2020

IET image process

View full text Add to dashboard Cite

In many scenarios, where videos are transmitted through bandwidth‐limited channels for subsequent semantic analytics, the choice of frame rates has to balance between bandwidth constraints and analytics performance. Faced with this practical challenge, this study focuses on enhancing object tracking at low frame rates and proposes a learning Interpolation for tracking framework. This framework embeds an implicit video frame interpolation sub‐network, which is concatenated and jointly trained with another object tracking sub‐network. Once a low frame‐rate video is an input, it is first mapped into a high frame‐rate latent video, based on which the tracker is learned. Novel strategies and loss functions are derived to ensure the effective end‐to‐end optimisation of the authors’ network. On several challenging benchmarks and settings, their method achieves a highly competitive tradeoff between frame rate and tracking accuracy. As is known, the implications of interpolation on semantic video analytics and tracking remain unexplored, and the authors expect their method to find many applications in mobile embedded vision, Internet of Things and edge computing.

show abstract

LaSOT: A High-quality Large-scale Single Object Tracking Benchmark

Cited by 289 publications

References 82 publications

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

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

Long-Term Visual Object Tracking via Continual Learning

End‐to‐end learning interpolation for object tracking in low frame‐rate video

Contact Info

Product

Resources

About