Towards More Flexible and Accurate Object Tracking with Natural Language: Algorithms and Benchmark

Wang, Xiao; Shu, Xiujun; Zhang, Zhipeng; Jiang, Bo; Wang, Yaowei; Wu, Feng

doi:10.1109/cvpr46437.2021.01355

Cited by 87 publications

(69 citation statements)

References 47 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We compare our SimTrack with state-of-the-art trackers on five datasets, including LaSOT [10], TNL2K [41], TrackingNet [30], UAV123 [29] and GOT-10k [19]. The compared trackers include SiamFC [2](ECCVW16), ATOM [11] (CVPR19), SiamRPN++ [42](CVPR19), DiMP [3](ICCV19), SiamFC++ [42] (AAAI20), Ocean [49](ECCV20), SiamBAN [9](CVPR20), SiamAtt [45] (CVPR 20), TransT [38](CVPR21), TrDiMP [39](CVPR21), KeepTrack [27](ICCV21), STARK-S [43](ICCV21), AutoMatch [48](ICCV21) and TransInMo * [15](ArXiV).…”

Section: State-of-the-art Comparisonsmentioning

confidence: 99%

“…Component-wise Analysis. To prove the efficiency of our method, we perform a component-wise analysis on the TNL2K [41] benchmark, as shown in Furthermore, the foveal window strategy brings an improvement of 0.8 point on AUC score in {. This shows using more detailed target patches at the beginning contributes to improving tracking accuracy.…”

Section: Ablation Study and Analysismentioning

confidence: 99%

“…• Extensive experiments on multiple datasets show the effectiveness of our method. Our SimTrack achieves state-of-the-art performances with 70.5% AUC on LaSOT [10], 55.6% AUC on TNL2K [41], 83.4% AUC on TrackingNet [30], 69.8% AO on GOT-10k [19] and 71.2% on UAV123 [29].…”

Section: Introductionmentioning

confidence: 99%

“…Performance comparisons with state-of-the-art trackers on the test set of LaSOT[10], TNL2K[41] and TrackingNet[30]. 'Net' denotes the backbone architecture; 'Size' means the size of search image, 'FLOPs' shows the computation costs of backbone and transformer head in different methods.…”

mentioning

confidence: 99%

See 3 more Smart Citations

Backbone is All Your Need: A Simplified Architecture for Visual Object Tracking

Chen¹,

Li²,

Bai³

et al. 2022

Preprint

View full text Add to dashboard Cite

Exploiting a general-purpose neural architecture to replace hand-wired designs or inductive biases has recently drawn extensive interest. However, existing tracking approaches rely on customized sub-modules and need prior knowledge for architecture selection, hindering the tracking development in a more general system. This paper presents a Simplified Tracking architecture (SimTrack) by leveraging a transformer backbone for joint feature extraction and interaction. Unlike existing Siamese trackers, we serialize the input images and concatenate them directly before the one-branch backbone. Feature interaction in the backbone helps to remove well-designed interaction modules and produce a more efficient and effective framework. To reduce the information loss from down-sampling in vision transformers, we further propose a foveal window strategy, providing more diverse input patches with acceptable computational costs. Our SimTrack improves the baseline with 2.5%/2.6% AUC gains on LaSOT/TNL2K and gets results competitive with other specialized tracking algorithms without bells and whistles.

show abstract

Section: State-of-the-art Comparisonsmentioning

confidence: 99%

Section: Ablation Study and Analysismentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

mentioning

confidence: 99%

See 2 more Smart Citations

Backbone is All Your Need: A Simplified Architecture for Visual Object Tracking

Chen¹,

Li²,

Bai³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…7: Obtain b t using {s i,j } and the post-processing algorithm in SiamCAR [11]; 8: end for Evaluation benchmarks and metrics: We evaluate the proposed tracker SiamLA on six challenging benchmarks, including GOT-10k [46], TrackingNet [48], LaSOT [49], TNL2K [50], OTB100 [51] and VOT2018 [52]. GOT-10k uses average overlap (AO) and success rate (SR) to evaluate the performance.…”

Section: A Implementation Detailsmentioning

confidence: 99%

Learning Localization-aware Target Confidence for Siamese Visual Tracking

Nie¹,

Han²,

He³

et al. 2022

Preprint

View full text Add to dashboard Cite

Siamese tracking paradigm has achieved great success, providing effective appearance discrimination and size estimation by the classification and regression. While such a paradigm typically optimizes the classification and regression independently, leading to task misalignment (accurate prediction boxes have no high target confidence scores). In this paper, to alleviate this misalignment, we propose a novel tracking paradigm, called SiamLA. Within this paradigm, a series of simple, yet effective localization-aware components are introduced, to generate localization-aware target confidence scores. Specifically, with the proposed localization-aware dynamic label (LADL) loss and localization-aware label smoothing (LALS) strategy, collaborative optimization between the classification and regression is achieved, enabling classification scores to be aware of location state, not just appearance similarity. Besides, we propose a separate localization branch, centered on a localization-aware feature aggregation (LAFA) module, to produce location quality scores to further modify the classification scores. Consequently, the resulting target confidence scores, are more discriminative for the location state, allowing accurate prediction boxes tend to be predicted as high scores. Extensive experiments are conducted on six challenging benchmarks, including GOT-10k, TrackingNet, LaSOT, TNL2K, OTB100 and VOT2018. Our SiamLA achieves state-of-the-art performance in terms of both accuracy and efficiency. Furthermore, a stability analysis reveals that our tracking paradigm is relatively stable, implying the paradigm is potential to real-world applications.

show abstract

Joint Feature Learning and Relation Modeling for Tracking: A One-Stream Framework

Chang

et al. 2022

Lecture Notes in Computer Science

179

View full text Add to dashboard Cite

Towards More Flexible and Accurate Object Tracking with Natural Language: Algorithms and Benchmark

Cited by 87 publications

References 47 publications

Backbone is All Your Need: A Simplified Architecture for Visual Object Tracking

Backbone is All Your Need: A Simplified Architecture for Visual Object Tracking

Learning Localization-aware Target Confidence for Siamese Visual Tracking

Joint Feature Learning and Relation Modeling for Tracking: A One-Stream Framework

Contact Info

Product

Resources

About