DeTracker: Motion-decoupled Vehicle Detection and Tracking in Unstabilized Satellite Videos
Not on the map. Why?
Abstract (arXiv)
Satellite videos provide continuous observations of surface dynamics but pose significant challenges for multi-object tracking (MOT), especially under unstabilized conditions where platform jitter and the weak appearance of tiny objects jointly degrade tracking performance. To address this problem, we propose DeTracker, a joint-detection-and-tracking framework tailored for unstabilized satellite videos. DeTracker introduces a task-driven Global-Local Motion Decoupling (GLMD) module to address the motion imbalance between dominant platform motion and weak target motion. It suppresses background-dominated motion via global semantic alignment at the feature level and captures target-specific motion through local refinement, improving trajectory stability and identity consistency. In addition, a Temporal Dependency Feature Pyramid (TDFP) module is developed to perform cross-frame temporal feature fusion, enhancing the continuity and discriminability of tiny-object representations. We further construct a new benchmark dataset, SDM-Car-SU, which simulates multi-directional and multi-speed platform motions to enable systematic evaluation of tracking robustness under varying motion perturbations. Extensive experiments on both simulated and real unstabilized satellite videos demonstrate that DeTracker significantly outperforms existing methods, achieving 61.1% MOTA on SDM-Car-SU and 45.3% MOTA on real satellite video data. The code and dataset will be publicly available at https://github.com/alex-chenjiajun/DeTracker.
Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 22 of its 73 references that are in the map. First posted on arXiv 14 Jan 2026.
Builds on
22 of 73 references.
References · 22
- 2019EfficientDet: Scalable and Efficient Object Detectioncited by 1,018
- 2017Simple Online and Realtime Tracking with a Deep Association Metriccited by 908
- 2021ByteTrack: Multi-Object Tracking by Associating Every Detection Boxcited by 812
- 2020FairMOT: On the Fairness of Detection and Re-Identification in Multiple Object Trackingcited by 489
- 2020Tracking Objects as Pointscited by 483
- 2017Deep Layer Aggregationcited by 436
- 2020Generalized Focal Loss: Learning Qualified and Distributed Bounding Boxes for Dense Object Detectioncited by 408
- 2022Observation-Centric SORT: Rethinking SORT for Robust Multi-Object Trackingcited by 345
- 2022StrongSORT: Make DeepSORT Great Againcited by 213
- 2015Delving Deeper into Convolutional Networks for Learning Video Representationscited by 146
- 2022AiATrack: Attention in Attention for Transformer Visual Trackingcited by 137
- 2021Learning Target Candidate Association to Keep Track of What Not to Trackcited by 131
10 more
- 2022Transforming Model Prediction for Trackingcited by 117
- 2023Robust Object Modeling for Visual Trackingcited by 80
- 2018A Twofold Siamese Network for Real-Time Object Trackingcited by 62
- 2024Multiple Object Tracking as ID Predictioncited by 58
- 2021Detecting and Tracking Small and Dense Moving Objects in Satellite Videos: A Benchmarkcited by 17
- 2018Deep Continuous Conditional Random Fields with Asymmetric Inter-object Constraints for Online Multi-object Trackingcited by 9
- 2019Error Bounded Foreground and Background Modeling for Moving Object Detection in Satellite Videoscited by 8
- 2024Highly Efficient and Unsupervised Framework for Moving Object Detection in Satellite Videoscited by 7
- 2023Single-Shot and Multi-Shot Feature Learning for Multi-Object Trackingcited by 5
- 2024SDM-Car: A Dataset for Small and Dim Moving Vehicles Detection in Satellite Videoscited by 2
Built on by
Citing papers per year: 2026: 1.
1 paper.