mlsanityData to 7 Oct 2026

Context-aware Visual Tracking with Joint Meta-updating

arXiv 2204.01513 ↗First posted 4 Apr 2022cs.CV cs.AIcited by 1

Not on the map. Why?

Abstract (arXiv)

Visual object tracking acts as a pivotal component in various emerging video applications. Despite the numerous developments in visual tracking, existing deep trackers are still likely to fail when tracking against objects with dramatic variation. These deep trackers usually do not perform online update or update single sub-branch of the tracking model, for which they cannot adapt to the appearance variation of objects. Efficient updating methods are therefore crucial for tracking while previous meta-updater optimizes trackers directly over parameter space, which is prone to over-fit even collapse on longer sequences. To address these issues, we propose a context-aware tracking model to optimize the tracker over the representation space, which jointly meta-update both branches by exploiting information along the whole sequence, such that it can avoid the over-fitting problem. First, we note that the embedded features of the localization branch and the box-estimation branch, focusing on the local and global information of the target, are effective complements to each other. Based on this insight, we devise a context-aggregation module to fuse information in historical frames, followed by a context-aware module to learn affinity vectors for both branches of the tracker. Besides, we develop a dedicated meta-learning scheme, on account of fast and stable updating with limited training samples. The proposed tracking method achieves an EAO score of 0.514 on VOT2018 with the speed of 40FPS, demonstrating its capability of improving the accuracy and robustness of the underlying tracker with little speed drop.

Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 44 of its 51 references that are in the map. First posted on arXiv 4 Apr 2022.

Builds on

44 of 51 references.

References · 44

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2017Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networkscited by 4,861
  3. 2016Fully-Convolutional Siamese Networks for Object Trackingcited by 566
  4. 2018Meta-Learning with Latent Embedding Optimizationcited by 476
  5. 2018SiamRPN++: Evolution of Siamese Visual Tracking with Very Deep Networkscited by 445
  6. 2018LaSOT: A High-quality Benchmark for Large-scale Single Object Trackingcited by 424
  7. 2021Transformer Trackingcited by 412
  8. 2019A Survey of Deep Learning Techniques for Autonomous Drivingcited by 401
  9. 2019Learning Discriminative Model Prediction for Trackingcited by 344
  10. 2018TrackingNet: A Large-Scale Dataset and Benchmark for Object Tracking in the Wildcited by 342
  11. 2018How to train your MAMLcited by 306
  12. 2018ATOM: Accurate Tracking by Overlap Maximizationcited by 285
32 more
  1. 2016ECO: Efficient Convolution Operators for Trackingcited by 239
  2. 2020Probabilistic Regression for Visual Trackingcited by 231
  3. 2015Learning Multi-Domain Convolutional Neural Networks for Visual Trackingcited by 209
  4. 2020Siamese Box Adaptive Network for Visual Trackingcited by 189
  5. 2018Distractor-aware Siamese Networks for Visual Object Trackingcited by 187
  6. 2019SiamFC++: Towards Robust and Accurate Visual Tracking with Target Estimation Guidelinescited by 177
  7. 2018Acquisition of Localization Confidence for Accurate Object Detectioncited by 174
  8. 2019Siam R-CNN: Visual Tracking by Re-Detectioncited by 174
  9. 2019Deeper and Wider Siamese Networks for Real-Time Visual Trackingcited by 156
  10. 2017Need for Speed: A Benchmark for Higher Frame Rate Object Trackingcited by 137
  11. 2020MAST: A Memory-Augmented Self-supervised Trackercited by 109
  12. 2021LambdaNetworks: Modeling Long-Range Interactions Without Attentioncited by 107
  13. 2020Deformable Siamese Attention Networks for Visual Object Trackingcited by 97
  14. 2019Learning the Model Update for Siamese Trackerscited by 75
  15. 2020Rotate to Attend: Convolutional Triplet Attention Modulecited by 75
  16. 2020Graph Attention Trackingcited by 70
  17. 2016Beyond Correlation Filters: Learning Continuous Convolution Operators for Visual Trackingcited by 66
  18. 2018Siamese Cascaded Region Proposal Networks for Real-Time Visual Trackingcited by 65
  19. 2020Tracking by Instance Detection: A Meta-Learning Approachcited by 61
  20. 2019Target-Aware Deep Trackingcited by 43
  21. 2018Unveiling the Power of Deep Trackingcited by 41
  22. 2019GradNet: Gradient-Guided Network for Visual Object Trackingcited by 41
  23. 2018Meta-Tracker: Fast and Robust Online Adaptation for Visual Object Trackerscited by 34
  24. 2017CREST: Convolutional Residual Learning for Visual Trackingcited by 29
  25. 2019ROAM: Recurrently Optimizing Tracking Modelcited by 28
  26. 2017A Survey on Content-Aware Video Analysis for Sportscited by 27
  27. 2020RPT: Learning Point Set Representation for Siamese Visual Trackingcited by 26
  28. 2019Learning Deep Multi-Level Similarity for Thermal Infrared Object Trackingcited by 19
  29. 2017DCFNet: Discriminant Correlation Filters Network for Visual Trackingcited by 11
  30. 2020Visual Tracking by TridentAlign and Context Embeddingcited by 9
  31. 2020Siamese Keypoint Prediction Network for Visual Object Trackingcited by 4
  32. 2020Contextualizing Enhances Gradient Based Meta Learningcited by 3

Built on by

Cited by year

Citing papers per year: 2025: 1.

1 paper.

  1. 2025Trajectory Prediction in Dynamic Object Tracking: A Critical Studycited by 0